Impact of Spin in the Abstracts of Articles Reporting Results of Randomized Controlled Trials in the Field of Cancer: The SPIIN Randomized Controlled Trial
Impact of Spin in the Abstracts of Articles Reporting Results of Randomized Controlled Trials in the Field of Cancer: The SPIIN Randomized Controlled Trial
复制标题
DOI：
10.1200/jco.2014.56.7503
复制
发表时间：
2014-12-20
影响因子：
45.3
通讯作者：
Ravaud, Philippe
中科院分区：
文献类型：
--
作者：
Boutron, Isabelle;Altman, Douglas G.;Ravaud, Philippe
PurposeWe旨在评估自旋（即，报告说服读者，实验性治疗的有益效果是大于所示的结果）对结果的解释摘要的随机对照试验（RCT）在cancer.MethodsWe进行了双臂，平行组RCT的影响。我们选择了一个样本的已发表的随机对照试验，统计学上不显着的主要结果和旋转的抽象结论。这些摘要的两个版本被使用了-原始的旋转和重写的版本没有旋转。参与者是报告RCT的文章的临床通讯作者、试验研究者和法国国家赠款的评审员。主要结果是临床医生对实验性治疗的有益效果的解释（0至10分）。参与者被蒙蔽研究hypothes.ResultsThree百名临床医生被随机分配使用基于Web的系统，150名临床医生评估的摘要与旋转和150评估的摘要没有旋转。对于有自旋的摘要，实验性处理被评为更有益（平均差异，0.71; 95%CI，0.07至1.35; P = 0.030），试验被评为不太严格（平均差异，-0.59; 95% CI，-1.13至0.05; P = .034），临床医生对阅读全文文章更感兴趣（平均差异，0.77; 95% CI，0.08至1.47; P = 0.029）。有没有统计学上的显着差异，在临床医生的评价的重要性的研究或需要运行另一个trial.ConclusionSpin摘要临床医生的解释的试验结果可能会产生影响。(C)2014年美国临床肿瘤学会
PurposeWe aimed to assess the impact of spin (ie, reporting to convince readers that the beneficial effect of the experimental treatment is greater than shown by the results) on the interpretation of results of abstracts of randomized controlled trials (RCTs) in the field of cancer.MethodsWe performed a two-arm, parallel-group RCT. We selected a sample of published RCTs with statistically nonsignificant primary outcome and with spin in the abstract conclusion. Two versions of these abstracts were used-the original with spin and a rewritten version without spin. Participants were clinician corresponding authors of articles reporting RCTs, investigators of trials, and reviewers of French national grants. The primary outcome was clinicians' interpretation of the beneficial effect of the experimental treatment (0 to 10 scale). Participants were blinded to study hypothesis.ResultsThree hundred clinicians were randomly assigned using a Web-based system; 150 clinicians assessed an abstract with spin and 150 assessed an abstract without spin. For abstracts with spin, the experimental treatment was rated as being more beneficial (mean difference, 0.71; 95% CI, 0.07 to 1.35; P = .030), the trial was rated as being less rigorous (mean difference, -0.59; 95% CI, -1.13 to 0.05; P = .034), and clinicians were more interested in reading the full-text article (mean difference, 0.77; 95% CI, 0.08 to 1.47; P = .029). There was no statistically significant difference in the clinicians' rating of the importance of the study or the need to run another trial.ConclusionSpin in abstracts can have an impact on clinicians' interpretation of the trial results. (C) 2014 by American Society of Clinical Oncology