Evaluation of Spin in the Abstracts of Systematic Reviews and Meta-Analyses Focused on Tinnitus.

Abstract

The objective was to investigate the prevalence of spin in abstracts of systematic reviews and meta-analyses covering the treatment of tinnitus. We hypothesized that spin would be present in these articles and a significant relationship would exist between spin usage and extracted study characteristics. Spin, the misrepresentation of study findings, can alter a clinician's interpretation of a study's results, potentially affecting patient care. Previous work demonstrates that spin is present in abstracts of randomized clinical trials. Using a cross-sectional analysis, we conducted a systematic search using MEDLINE and Embase databases on June 2, 2020, for systematic reviews focused on tinnitus treatment. Investigators performed screening and data extraction in a masked, duplicate fashion. Forty systematic reviews met inclusion criteria, and spin was identified in four of them. Spin in abstracts most frequently occurred when conclusions claimed the beneficial effect of the experimental treatment despite high risk of bias in primary studies (n = 3). The other form of spin found was the conclusion claims safety based on nonstatistically significant results with a wide confidence interval (n = 1). There was no significant association between spin and any of our extracted study characteristics. Spin was observed in 10% of abstracts of systematic reviews and meta-analyses covering the treatment of tinnitus. Although this percentage may be small, we recommend that medical journals provide a more detailed framework for abstract structure and require the inclusion of risk of bias assessment results in abstracts to prevent the incorporation of spin.