Loading...
Authors of journal articles often frame discussions to advance particular perspectives or agendas. To examine this issue, researchers searched PubMed and identified all randomized controlled trials (RCTs) published in December 2006 and selected the 72 RCTs with clearly identified primary outcomes that were not statistically significant (P≥0.05). The researchers analyzed each article for presence of “spin,” using their own classification scheme.
Spin was identified in 18% of article titles; in 38% and 58% of abstract results and conclusion sections, respectively; and in 29%, 43%, and 50% of main-text results, discussion, and conclusion sections, respectively. The level of spin in conclusion sections was considered “high” in about 30% of artic…