Universidade Positivo, Department of Dentisty, Curitiba, PR, Brazil.
Universidade Estadual de Ponta Grossa - UEPG, Department of Dentistry, PR, Brazil.
Braz Oral Res. 2022 Feb 9;36:e017. doi: 10.1590/1807-3107bor-2022.vol36.0017. eCollection 2022.
Appropriate research reports are important to facilitate the evaluation of studies and the decision-making by dentists and policymakers. This meta-research study assessed the conformity of randomized clinical trials (RCTs) on atraumatic restorative treatment (ART) restorations with the CONSORT recommendations and their risk of bias (RoB). Cochrane Library, MEDLINE, BBO, LILACS, Scopus, and Web of Science databases were searched from April 2019 to June 2021 for RCTs that assessed the longevity of ART restorations in children. A specific tool was used to assess adherence to the CONSORT recommendations; RoB was evaluated with the Cochrane risk-of-bias tool. Descriptive analyses included the number of studies by journal, follow-up period, country, and quality assessments. A total of 2,181 papers were retrieved and 36 of them were analyzed qualitatively. The overall CONSORT mean score (CONms) was 22.52 ± 6.17 out of 32 points. The best described items were intervention and outcomes, whereas allocation concealment was described in only 22% of the papers. Significant differences in CONms were detected in the analysis by country and publication dates. High CONms were observed in recently published papers (26.7 ± 3.1) when compared to first ART studies (18.1 ± 4.6; p < 0.001). RoB was low in four studies, unclear in 11, and high in 21. Adherence of the papers to the CONSORT recommendations was not fully achieved and most of the papers had unclear and high RoB (PROSPERO registration #CRD42020201460).
适当的研究报告对于促进研究评估和牙医及政策制定者的决策至关重要。本元研究评估了关于无创伤性修复治疗（ART）修复体的随机临床试验（RCT）与 CONSORT 建议的一致性及其偏倚风险（RoB）。从 2019 年 4 月至 2021 年 6 月，在 Cochrane 图书馆、MEDLINE、BBO、LILACS、Scopus 和 Web of Science 数据库中搜索评估儿童 ART 修复体耐久性的 RCT 报告。使用特定工具评估对 CONSORT 建议的遵守情况；使用 Cochrane RoB 工具评估 RoB。描述性分析包括期刊、随访期、国家和质量评估的研究数量。共检索到 2181 篇论文，其中 36 篇进行了定性分析。总体 CONSORT 平均得分（CONms）为 32 分中的 22.52 ± 6.17 分。干预和结局描述得最好，而仅 22%的论文描述了分配隐匿。按国家和出版日期进行分析时，CONms 存在显著差异。与最初的 ART 研究（18.1 ± 4.6；p < 0.001）相比，最近发表的论文（26.7 ± 3.1）的 CONms 较高。4 项研究的 RoB 较低，11 项研究的 RoB 不明确，21 项研究的 RoB 较高。这些论文对 CONSORT 建议的遵守程度并未完全达到，并且大多数论文的 RoB 不明确和较高（PROSPERO 注册号：CRD42020201460）。
