Isabelle Boutron1, Douglas G Altman2, Sally Hopewell2, Francisco Vera-Badillo2, Ian Tannock2, Philippe Ravaud2. 1. Isabelle Boutron, Sally Hopewell, and Philippe Ravaud, Methods of Therapeutic Evaluation of Chronic Diseases Team, Epidemiology and Biostatistics Sorbonne Paris Cité Research Centre, Unite Mixte de Recherche 1153, L'Institut National de la Santé et de la Recherche Médicale; Isabelle Boutron and Philippe Ravaud, Assistance Publique des Hôpitaux de Paris, Hôpital Hôtel Dieu; Isabelle Boutron and Philippe Ravaud, Paris Descartes University, Sorbonne Paris Cité; Isabelle Boutron, Sally Hopewell, and Philippe Ravaud, French Cochrane Centre, Paris, France; Douglas G. Altman and Sally Hopewell, Centre for Statistics in Medicine, University of Oxford, Oxford, United Kingdom; Francisco Vera-Badillo and Ian Tannock, Princess Margaret Cancer Centre and University of Toronto, Toronto, Ontario, Canada; Philippe Ravaud, Columbia University Mailman School of Public Health, New York, NY. isabelle.boutron@htd.aphp.fr. 2. Isabelle Boutron, Sally Hopewell, and Philippe Ravaud, Methods of Therapeutic Evaluation of Chronic Diseases Team, Epidemiology and Biostatistics Sorbonne Paris Cité Research Centre, Unite Mixte de Recherche 1153, L'Institut National de la Santé et de la Recherche Médicale; Isabelle Boutron and Philippe Ravaud, Assistance Publique des Hôpitaux de Paris, Hôpital Hôtel Dieu; Isabelle Boutron and Philippe Ravaud, Paris Descartes University, Sorbonne Paris Cité; Isabelle Boutron, Sally Hopewell, and Philippe Ravaud, French Cochrane Centre, Paris, France; Douglas G. Altman and Sally Hopewell, Centre for Statistics in Medicine, University of Oxford, Oxford, United Kingdom; Francisco Vera-Badillo and Ian Tannock, Princess Margaret Cancer Centre and University of Toronto, Toronto, Ontario, Canada; Philippe Ravaud, Columbia University Mailman School of Public Health, New York, NY.
Abstract
PURPOSE: We 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. METHODS: We 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. RESULTS: Three 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. CONCLUSION: Spin in abstracts can have an impact on clinicians' interpretation of the trial results.
PURPOSE: We 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. METHODS: We 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. RESULTS: Three 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. CONCLUSION: Spin in abstracts can have an impact on clinicians' interpretation of the trial results.
Authors: Emma Mead; Tamara Brown; Karen Rees; Liane B Azevedo; Victoria Whittaker; Dan Jones; Joan Olajide; Giulia M Mainardi; Eva Corpeleijn; Claire O'Malley; Elizabeth Beardsmore; Lena Al-Khudairy; Louise Baur; Maria-Inti Metzendorf; Alessandro Demaio; Louisa J Ells Journal: Cochrane Database Syst Rev Date: 2017-06-22