Abstract ‘Spin’ Affects Interpretation of Trial Outcome
In the SPIIN study reported in the Journal of Clinical Oncology, Boutron et al found that spin in the abstract of an article reporting results of a randomized controlled trial in cancer increased clinician rating of the benefit of the experimental treatment in the trial.
Study Details
In the study, 300 clinicians mainly from Europe (n = 169), the United States (n = 77), and Canada (n = 20) were randomly assigned to receive an abstract from a published trial with statistically nonsignificant primary outcome without spin in the abstract (n = 150) or with the abstract rewritten to contain spin (n = 150). The clinicians were corresponding authors of articles reporting randomized controlled trials, investigators in trials, and reviewers of French national grants.
The primary outcome was clinician interpretation of the beneficial effect of the experimental treatment on a 0 to 10 scale, elicited by the question “Based on this abstract, do you think treatment A would be beneficial to patients? (Scale, 0 [very unlikely] to 10 [very likely]).” Other items, also using the 0 to 10 scale, were: “Rate the overall rigor of the study methodology”; “Rate the importance of the study”; “Are you interested in reading the full-text article for the study described in this abstract?”; and “Do you think it would be interesting to run another trial evaluating this treatment?”
Examples of spin included: no identification of primary outcome in methods section of abstract; incomplete reporting of results for primary outcome; selective reporting of statistically significant secondary outcomes; no or incomplete reporting of safety data; focus on statistically significant subgroup or secondary analysis; linguistic spin; and conclusions with spin.
Outcomes
For abstracts with spin vs those without, the experimental treatment was rated as more beneficial (mean score 3.6 vs 2.9, difference = 0.71, P = .03), the trial was rated as being less rigorous (4.5 vs 5.1, difference = −0.59, P = .034), and there was greater interest in reading the full-length article (5.1 vs 4.3, difference = 0.77, P = .029). There was no significant difference in rating of importance of study (4.6 vs 4.9, difference = −0.38, P = .18) or rating of interest in conducting another trial to evaluate the experimental treatment (4.8 vs 4.2, difference = 0.64, P = .061)
The authors concluded: “Spin in abstracts can have an impact on clinicians’ interpretation of the trial results…. It is important to raise the awareness of this problem within the scientific community to limit potential harm.”
Isabelle Boutron, MD, of Centre d’Épidémiologie Clinique, Hôpital Hôtel Dieu, Paris, is the corresponding author for the Journal of Clinical Oncology article.
The study was supported by the French Ministry of Health, National Institute for Cancer Research, Fondation Pour la Recherche Médicale, and Cancer Research UK. Ian Tannock, MD, PhD, reported research funding from Sanofi.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
