The threat of serious outcome reporting bias in randomized controlled trials on acute ischemic stroke to evidence synthesis: a meta-epidemiological study

Na Zhang; Youlin Long; Xinyao Wang; Xinyi Wang; Qiong Guo; Zhengchi Li; Liang Du

doi:10.21037/cdt-2025-212

The threat of serious outcome reporting bias in randomized controlled trials on acute ischemic stroke to evidence synthesis: a meta-epidemiological study

Cardiovasc Diagn Ther. 2025 Dec 31;15(6):1182-1193. doi: 10.21037/cdt-2025-212. Epub 2025 Dec 19.

Authors

Na Zhang^#^{1

2

3}, Youlin Long^#^{2

3}, Xinyao Wang^{2

3}, Xinyi Wang³, Qiong Guo^{3

4}, Zhengchi Li³, Liang Du^{2

3}

Affiliations

¹ General Practice Medical Center, West China Hospital, Sichuan University, Chengdu, China.
² Chinese Evidence-Based Medicine Centre, West China Hospital, Sichuan University, Chengdu, China.
³ Innovation Institute for Integration of Medicine and Engineering, West China Hospital, Sichuan University, Chengdu, China.
⁴ West China Medical Publishers, West China Hospital, Sichuan University, Chengdu, China.

^# Contributed equally.

Abstract

Background: Stroke is the second-leading cause of death and the third-leading cause of disability, with acute ischemic stroke (AIS) being the most serious subtype. Systematic reviews of randomized controlled trials (RCTs) for AIS play a crucial role in formulating clinical guidelines and health policies. However, potential outcome reporting bias (ORB) in RCTs may skew the analytical results of systematic reviews and ultimately lead to suboptimal medical decisions. This study was conducted to investigate the prevalence and possible influencing factors of ORB in RCTs included in systematic reviews of AIS and to correct ORB at the level of systematic reviews.

Methods: A systematic literature search was conducted across three databases to retrieve subject headings and text terms related to AIS, RCTs, and systematic reviews, with the aim of identifying AIS-related systematic reviews published in 2022. ORB in trials was employed to assess the risk of ORB in RCTs, and multivariate logistic regression was used to identify the possible ORB-related factors, including registration, country, quality of journal, funding, sample size, and type of control. The correcting for ORB model was used to correct ORB evidence synthesis results.

Results: A total of 33 systematic reviews and 287 nonduplicate RCTs were included in this study. ORB was suspected in 138 (48.08%) of these RCTs. Statistically significant outcomes were more likely to be reported than were nonsignificant ones [relative risk (RR) =3.18; 95% confidence interval (CI): 2.77-3.64]. The potential factors associated with ORB were unregistered status [odds ratio (OR) =4.87; 95% CI: 1.93-12.28] and sample sizes smaller than 100 (OR =2.57; 95% CI: 1.30-5.10). The corrected results indicated that 31.58% of the therapeutic effects were overestimated due to reversal and that 16.67% of adverse reactions were underestimated due to reversal. Among outcomes without reversal, 56.52% of the effect sizes and 60.87% of the P values exceeded the clinically acceptable range.

Conclusions: The presence of ORB within the field of AIS poses a serious threat to the reliability of evidence synthesized in systematic reviews. In the future, healthcare practitioners and decision-makers should adopt a critical perspective when applying seemingly favorable results in clinical practice.

Keywords: Acute ischemic stroke (AIS); meta-epidemiology; selection bias; systematic review.

Publication types

Review