Extent and prevalence of spin in randomized controlled trials in dentistry.

OBJECTIVE

Spin, the misinterpretation of non-significant study results, can lead to erroneous conclusions about the effectiveness of interventions. It was the aim of this study to examine the presence and prevalence of spin in the dental literature and to identify potential associations with trials' characteristics.

METHODS

PubMed was searched for randomized controlled trials (RCTs) published between February 12,018 and January 31, 2019. Two reviewers screened the selected articles and evaluated them for spin using a standardized data collection form. Data on the selected RCTs and spin strategy, if present, were recorded for each study. Descriptive statistics and frequencies of various types of spin in the abstract and the main text were calculated and associations between spin and study characteristics were examined.

RESULTS

Forty-seven RCTs were included in the final sample. Over 60 % of the articles presented some form of spin and 19.1 % presented more than one types of spin in the abstract. More than one types of spin were detected in more than half of the studies evaluated (53.2 %), whereas 78.7 % of the studies screened presented some type of spin in the main text. No statistically significant associations were found between any of the study characteristics and spin or recommendations. There was a discrepancy in the presence of spin between abstracts and in the main text (p = 0.006).

CONCLUSION

Dental RCTs' are not immune to misleading interpretation and both authors and readers should report and evaluate trial results based on the data at hand without unfounded extrapolations.

CLINICAL SIGNIFICANCE

Spin, the conscious or subconscious misinterpretation of non-significant study results, can be misleading when applying trials' results in everyday clinical practice. Thus, researchers should abstain from including it in their trials and readers should identify it when evaluating a study.