Jonas Ranstam's website
3. Spin in medical research
A manuscript usually starts with an introduction explaining why a study is performed and a detailed description of the conceptual research question behind the study's specific aim. The manuscript then describes the patients that provide data and how the data are collected and evaluated. Other sections present the evaluation results and a discussion of the strengths and weaknesses of the study. All this leads to the authors' conclusion regarding the research question explained in the introduction.
Ideally, the research question can be answered by the used data and methods, and the authors' conclusions are consistent with the analysis results and the limitations dictated by the practical circumstances, i.e. no unsubstantiated claims are made. In practice, however, many manuscripts have no precise aim. The validity of the results from the analysed data is unclear, and there is either no conclusion given, or no direct link exists between the actual findings and the authors' conclusion.
Some of these problems can probably be explained by mistakes caused by insufficient experience, but it is crucial to realise that the consequences of the errors are not random. Many authors tend to exaggerate both the empirical support for their findings and the findings' scientific value. Some of the problems can also be explained by spin, i.e. reporting practices that intentionally mislead readers to view the findings in a more favourable light. This phenomenon is more common than could be expected.
One systematic review shows, for example, that spin occurred in about half of all press releases from randomised trials (Yavchitz et al. Misrepresentation of Randomized Controlled Trials in Press Releases and News Coverage: A Cohort Study. PLoS Med 2012;9:e1001308.) and another systematic review (Chiu et al. ‘Spin’ in published biomedical literature: A methodological systematic review. PLoS Biol 2017;15:e2002173) could not find any difference in prevalence between different types of studies.[Return]