Selective Outcome Reporting Bias in Randomized Controlled Trials on Dental Caries in Children and Adolescents: A Meta-Research Study.

INTRODUCTION

Selective outcome reporting (SOR) is a bias that can occur in randomized controlled trials (RCTs), defined as the alteration or omission of primary outcome in the publication compared to the original protocol. Researchers may modify outcomes to highlight statistically significant results. This study aimed to assess the prevalence of SOR in RCTs related to dental caries in children and adolescents.

METHODS

We conducted a search on <ext-link ext-link-type="uri" xlink:href="http://ClinicalTrials.gov" xmlns:xlink="http://www.w3.org/1999/xlink">ClinicalTrials.gov</ext-link> and the International Clinical Trials Registry Platform (ICTRP), using a comprehensive search strategy with terms related to pediatric dentistry and dental caries, up to February 2023. Two independent reviewers included trials with two or more arms focusing on dental caries in pediatric dentistry. Registrations that did not result in at least one published article were excluded. Data on the characteristics and outcomes from the protocols and corresponding publications were extracted. The primary outcome was the prevalence of SOR in the included RCTs. A chi-square test, with a significance level of 5%, was used to assess the association between SOR and pre-specified variables, which was the secondary outcome.

RESULTS

A total of 175 protocols and their corresponding publications were included. SOR was identified in 58.9% (n = 103) of the studies, with 41.1% (n = 72) showing discrepancies in the primary outcome's time frame. Retrospective registrations accounted for 73.7% of the studies. SOR was significantly associated with discrepancies in the follow-up period (p < 0.001) and with study design type (parallel assignment, split-mouth, cluster RCT, and no information) (p = 0.048).

CONCLUSIONS

The high prevalence of SOR in dental caries RCTs in pediatric dentistry highlights the need for attention to this issue. Ensuring transparency in the research process requires implementing an appropriate pre-registered protocol, disclosing deviations from it, and enabling stakeholders to compare the protocol with the published outcomes which can help reduce research waste.