Review
A systematic review considering the risk of bias in orthodontic RCTs over 55 years
| Title: | A systematic review considering the risk of bias in orthodontic RCTs over 55 years |
|---|---|
| Authors: | Petrén, Sofia, Bondemark, Lars, Sonesson, Mikael, Paulsson, Liselotte |
| Source: | European Journal of Orthodontics. 47(6) |
| Subject Terms: | Humans, Randomized Controlled Trials as Topic / standards, Orthodontics, Bias, Research Design / standards, RCTs, risk of bias, systematic review |
| Description: | BACKGROUND: When assessing the effectiveness of interventions, randomized controlled trials (RCTs) are considered to generate the highest level of evidence. The CONSORT 2010 recommendations promote clear and transparent reporting of RCTs. However, to perform a complete analysis of methodological errors and risk of bias that RCTs may be subjected to, the Risk of Bias tool 2 (RoB 2) has been constructed.OBJECTIVES: The aim of this systematic review was to assess the last 55 years of changes in methodological quality of orthodontic RCTs by using the RoB 2 tool.SEARCH METHODS: The MEDLINE via Entrez PubMed, Web of Science, and Cochrane library databases were searched for orthodontic RCT publications from 1 January 1968 to 31 December 2024.SELECTION CRITERIA: All RCTs on humans in the field of orthodontics were included. Non-randomized trials, animal studies, orthognathic surgery, syndrome patients, CLP patients, sleep apnea, and in vitro studies were excluded.DATA COLLECTION AND ANALYSIS: The quality assessment of the studies was conducted using the Cochrane Risk of Bias Tool (RoB2). The RCTs were divided into three time periods, i.e. before CONSORT 2010, from 2011 to 2016, and from 2017 to 2024.RESULTS: A total of 3135 RCTs were identified, and after excluding studies not fulfilling the criteria, 1231 RCTs were included in the quality assessment. Publications in the later or latest time-period had a larger number with low risk of bias than early ones. However, significant room for improvement remained since, in the latest time period there were relatively many RCTs with high risk of bias (67.6%). The main factors to high bias were unclear or missing information about 'who generated the random allocation sequence and enrolled participants' (selection bias), whether 'ITT or intention-to-treat' was used (attrition bias) as well as omitting reporting of 'all important harms or unintended effects' (other bias). Another factor was that baseline characteristics were missing (selection bias).LIMITATIONS: The RoB2 tool is complex, and it requires trained individuals to use the tool.CONCLUSIONS: There remains a need to enhance the quality of RCTs in orthodontics. To reduce the risk of bias, researchers should become well-acquainted with RoB2 before developing research protocols and plans. Specifically, use the 'signaling questions of RoB2' as extensively as possible to aid in designing the plans and protocols concerning the trial's conduct and progression.REGISTRATION: PROSPERO: CRD42023390206. |
| File Description: | electronic |
| Access URL: | https://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-80163 https://doi.org/10.1093/ejo/cjaf083 |
| Database: | SwePub |
| ISSN: | 01415387 14602210 |
|---|---|
| DOI: | 10.1093/ejo/cjaf083 |