Prevalence and Severity of Spin in Systematic Reviews and Meta-Analyses Evaluating Treatment for Shoulder Instability.

Abstract

Background: Shoulder instability is a common and clinically significant condition, particularly among young and active populations, and represents a frequent indication for both nonoperative and operative intervention. As the volume of clinical research and comparative surgical literature continues to grow, systematic reviews and meta-analyses are increasingly relied upon to guide treatment selection, patient counseling, and return-to-sport expectations. In this setting, biased reporting within evidence syntheses may meaningfully influence clinical decision-making. The purpose of this study was to evaluate the prevalence and types of reporting bias (spin) in systematic reviews and meta-analyses addressing shoulder instability and to determine whether spin is associated with methodological rigor.

Methods: This systematic review was prospectively registered with PROSPERO (CRD420261399489). Following PRISMA guidelines, PubMed, Scopus, and Embase were searched for systematic reviews and meta-analyses focused on shoulder instability in December 2025. Eligible studies were evaluated for 15 predefined types of spin using the Yavchitz taxonomy, and methodological quality was assessed using AMSTAR 2. Study characteristics included PRISMA adherence, protocol preregistration, level of evidence, publication year, and journal impact factor. Associations between spin and study characteristics were analyzed using Chi-square, Mann-Whitney U, Kruskal-Wallis, and Spearman rank correlation tests.

Results: A total of 126 studies were included. Spin was identified in 86 studies (68.3%), most commonly involving misleading reporting (55.6%) and misleading interpretation (38.1%). Most studies were rated as critically low confidence on AMSTAR 2 (73.0%). The presence and burden of spin were significantly associated with lower numerical AMSTAR 2 scores (p < 0.01). Level of evidence was associated with spin burden but not spin presence. Journal impact factor and year of publication were not associated with spin.

Conclusion: Reporting spin is highly prevalent in systematic reviews and meta-analyses addressing shoulder instability and is strongly associated with lower methodological quality. Given the high prevalence of shoulder instability and the reliance on evidence syntheses to inform treatment decisions, patient counseling, and expectations regarding recurrence and return to sport, biased reporting may contribute to misinterpretation of treatment effectiveness and inappropriate clinical conclusions. These findings underscore the need for greater methodological rigor, transparency, and critical appraisal of evidence syntheses to support informed, evidence-based care for patients with shoulder instability.