Spin in Randomized Controlled Trials of Pharmacology in COVID-19: A Systematic Review
Objectives: Spin, defined as the misrepresentation of the results of a study, could negate the validity of scientific findings. We aimed to evaluate spin in early published RCTs on pharmacotherapy for COVID-19 with nonsignificant primary outcomes, explore the manifestation of spin, and identify the factors affecting spin in COVID-19 RCTs.
Methods: A systematic review was performed through a search of RCTs from PubMed/Medline, National Institutes of Health, EMBASE, Cochrane, and Web of Science. RCTs on pharmacotherapy for COVID-19 with nonsignificant primary outcomes published in 2020 were included.
Results: Of the included articles, 21 abstracts (33.9%) and 28 main texts (45.2%) were found to contain spin in at least one section. In the conclusion section, other spin strategies beautifying their findings that were not included in the abstract were found in the main texts. Although specialized medical journals and the small sample size of the included RCTs were common factors that were associated with higher odds of spin in both abstracts in univariate analysis, they were not significant predictors of spin in the multivariate analysis. More factors influencing the level of spin were found in abstracts than in the main texts, but most of the levels of spin in abstracts were comparable to those in the main texts.
Conclusion: Although common factors that affected the manifestation of spin in the main texts and abstracts were the sample size and type of journal, further research to determine multicollinearity between significant factors and the manifestation of spin is required.
History
Refereed
- Yes
Publication title
Accountability in researchISSN
0898-9621External DOI
Publisher
Taylor and Francis GroupFile version
- Accepted version
Item sub-type
ArticleAffiliated with
- School of Psychology and Sport Science Outputs