Spin in the abstracts of randomized controlled trials of nurse-led care: A cross-sectional study
Dongguang Wang,Lian Wang,Xiang Tong,Sitong Liu,Hong Fan,Yonggang Zhang
DOI: https://doi.org/10.1016/j.ijnurstu.2023.104543
IF: 8.1
2023-06-02
International Journal of Nursing Studies
Abstract:Background Nurse-led models of care had been reported as effective intervention approaches for improving health management and reducing hospitalizations of target patients in a number of studies. However, the reporting quality studies in the field varied and there was a lack of literature evaluation. Objective The objective of the study was to assess the reporting quality and spin of abstracts of published randomized controlled trials which had statistically not significant primary outcomes. Moreover, potential factors associated with the presence of spin was also assessed. Methods Studies on nurse-led care were retrieved from PubMed from January 1st, 2017, to December 31st, 2021. Only randomized controlled trials with statistically not significant primary outcomes were included. Study screening and data extraction were carried out by two reviewers independently. The reporting quality of each abstract was evaluated by the Consolidated Standards of Reporting Trials statement, and spin strategies were analyzed using a pre-designed assessment form. Potential predictors for the presence of spin were analyzed by multivariate logistic regressions. Results The overall reporting quality of the included 75 randomized controlled trial abstracts was not satisfying, with a median score of 16-item Consolidated Standards of Reporting Trials statement at 6 (IQR 5, 8). Forty abstracts used at least one spin strategy in abstracts. Among them, 18 (45.0%) used spin strategies in the result section and 39 (97.5%) had spin in the conclusion section. The most common spin strategy identified in abstracts was focusing on statistically significant secondary outcomes (12/40, 30.0%) in the result section and claiming benefit with no consideration of statistically not significant results for the primary outcomes (32/40, 80.0%) in the conclusion section. Based on the definition, 29 (72.5%) abstracts were assessed to have high level of spin in the conclusions of abstracts. By multivariate logistic regression analyses, it was found that only geographic origin (reference: studies from Asian countries, OR = 0.118, 95%CI 0.027 to 0.511, P = 0.004) and the Consolidated Standards of Reporting Trials statement score (reference: lower score, OR = 0.625, 95%CI 0.470 to 0.829, P = 0.001) were significantly associated with the presence of spin in abstracts. Conclusion Among the randomized controlled trials with statistically not significant primary outcomes in the field of nurse-led care, the reporting quality of abstracts needs to be improved. Trials from Asian countries and with lower Consolidated Standards of Reporting Trials statement scores are more likely to present spin in abstracts. Findings reported in the result and conclusion sections of these abstracts need to be interpreted with caution.
nursing