Evaluation of spin in the abstracts of systematic reviews and meta-analyses focused on the treatment of obesity.

Overview

OBJECTIVE: Spin, i.e., the misrepresentation of research findings, has the potential to affect patient care. Evidence suggests that spin is prevalent in obesity randomized controlled trials. Therefore, the primary objective of this study was to evaluate spin in abstracts of systematic reviews covering obesity treatments. METHODS: MEDLINE and Embase were searched to retrieve systematic reviews on obesity treatments. Each systematic review abstract was inspected for the nine most severe types of spin, i.e., the misrepresentation of study findings by exaggeration or omission, regardless of intentionality. Screening and data extraction occurred in a masked, triplicate fashion. Methodological quality was determined using A MeaSurement Tool to Assess systematic Reviews (AMSTAR-2). RESULTS: Spin was identified in 20 (out of 200, 10%) abstracts, with spin type 5 (claiming efficacy despite high risk of bias among primary studies) being most common (11/200, 5.5%). Spin types 2 and 7, both related to unsupported efficacy claims, were not found. No associations were found between spin and extracted study characteristics. The methodological quality of the sample was rated as follows: critically low (23.0%), low (13.5%), moderate (60.5%), and high (3%). CONCLUSIONS: Although these findings demonstrate a low proportion of spin in the abstracts of systematic reviews for obesity treatment; increased preventive measures may further reduce its presence.