Heikkinen, J., Jokihaara, J., Das De, S., Jaatinen, K., Buchbinder, R., & Karjalainen, T. (2022). Bias in hand surgical randomized controlled trials: Systematic review and meta-epidemiological study. The Journal of Hand Surgery, 47(6), 526–533. https://doi.org/10.1016/j.jhsa.2022.01.027
Bias in hand surgical randomized controlled trials : systematic review and meta-epidemiological study
|Author:||Heikkinen, Juuso1; Jokihaara, Jarkko2,3; Das De, Soumen4;|
1Department of Surgery, Division of Orthopedic and Trauma Surgery, Oulu University Hospital, Medical Research Center, University of Oulu, Oulu, Finland
2Department of Hand Surgery, Tampere University Hospital, Tampere, Finland
3Faculty of Medicine and Life Sciences, Tampere University, Tampere, Finland
4Department of Hand & Reconstructive Microsurgery, National University Health System, Singapore
5University of Jyväskylä, Jyväskylä, Finland
6Monash Department of Clinical Epidemiology, Cabrini Institut, Australia, Melbourne
7Department of Epidemiology and Preventive Medicine, School of Public Health & Preventive Medicine, Monash University, Australia, Melbourne
8Clinical Trials Unit, Warwick Medical School, Warwick University, Coventry, Great Britain
9Unit of Hand Surgery, Department of Surgery, Central Finland Central Hospital, Jyväskylä, Finland
|Online Access:||PDF Full Text (PDF, 1.7 MB)|
|Persistent link:|| http://urn.fi/urn:nbn:fi-fe2023030730369
|Publish Date:|| 2023-03-22
Purpose: Inappropriately reported or conducted studies may decrease the quality of care due to under- or overestimation of the benefits or harms of interventions. Our aim was to evaluate how often hand surgical randomized controlled trials (RCTs) use and report adequate methods to ensure internal validity, and whether inadequate reporting or methods are associated with the magnitude of treatment effect estimates.
Methods: Data Sources were the Cochrane Central Register of Controlled Trials, MEDLINE, and Embase databases until November 2020. We included published RCTs investigating the effects of any surgical intervention in the hand and wrist region. We assessed internal validity using the Cochrane Risk of Bias (RoB) tool for 6 domains: selection, performance, detection, attrition, selective reporting, and “other” bias. We extracted the primary outcome and calculated the effect size for each study. We used mixed-effect meta-regression to assess whether the RoB modified the magnitude of the effects.
Results: For 207 assessed trials, the RoB was unclear or high for 72% in selection, 93% in performance, 88% in detection, 25% in attrition, 22% in selective reporting, and 34% in the “other” bias domain. Trials with a high or unclear risk of selection bias yielded 0.28 standardized mean difference (95% confidence interval, 0.02–0.55) larger effect sizes compared to studies with a low risk. Risks of bias for other domains did not modify the intervention effects. The risk for selection bias declined over time: the odds ratio for a high or unclear RoB was 0.90 (95% confidence interval, 0.85–0.95) per additional year of publication.
Conclusions: The internal validity and credibility of hand surgical RCTs can be improved by using established methods to achieve true randomization, blinding of the participants and study personnel, publishing the trial protocol and avoiding selective reporting of the outcomes, and reporting the trial as recommended in the Consolidated Standards of Reporting Trials statement.
Clinical relevance: Clinicians should be aware that RCTs that do not use or report proper randomization and allocation concealment may overestimate the treatment effects.
Journal of hand surgery
|Pages:||526 - 533|
|Type of Publication:||
A1 Journal article – refereed
|Field of Science:||
3126 Surgery, anesthesiology, intensive care, radiology
© 2022. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/