An Evaluation of Spin in Lower Extremity Joint Trials

Abstract

Background: Spin is a specific type of reporting bias that misrepresents data and results within randomized controlled trials (RCTs). Because spin may provide a surgeon with an inaccurate representation of trial results, thus misconstruing the surgeons' interpretation thereof and possibly negatively affecting patient care, it is important that spin is identified within publications. The primary goal of our study was to determine the prevalence of spin found within the abstracts of lower extremity joint trials.

Methods: Using Google Scholar's H-5 index, we selected the top 20 journals in the orthopedic surgery category. We then conducted a PubMed search on July 2nd, 2018 using the advanced search feature, encompassing all RCTs published in these journals from January 1, 2016 to January 1, 2017. Spin was evaluated using a standardized protocol, using a previously published protocol on the Open Science Network.

Results: Our final sample consisted of 46 trials published in 9 of the top 20 orthopedic surgery journals. Spin was found in 27 (58.7%) of the 46 abstracts. Evidence of spin in the abstract results was found in 19 (41.3%) of the 46 articles, and spin in abstract conclusions was found in 15 (32.6%) of the 46 articles.

Conclusions: Our study found that a significant number of lower extremity joint RCTs contain one or more form of spin in either their abstract results, conclusions, or both. In addition, our investigation revealed that a sizable portion of these lower extremity orthopedic joint RCTs are not registered or do not report their registration, and funding sources are also underreported.

Keywords: arthroplasty; joint trials; publication bias; reporting bias; spin.