新生儿试验主要结局报告异质性与不足
该研究评估2015至2020年发表的36项新生儿试验(每组≥100例)的主要结局报告完整性,发现报告水平差异大,所有试验完整报告了测量领域、统计方法和受试者流程,但仅28%完整报告最小重要差异,24%报告缺失数据,66%报告结局评估者盲法,42%报告多重性处理。结论认为新生儿试验主要结局报告常缺乏可解释性和证据合成所需的关键信息。
要点
- 新生儿试验主要结局报告完整性变异大,部分关键信息缺失普遍。
- 仅28%试验完整报告最小重要差异,24%报告数据缺失,42%报告多重性处理。
- 所有试验完整报告了主要结局测量领域、统计方法和受试者流程。
核心结论
- 新生儿试验主要结局报告常缺乏结果可解释性、知识合成和证据决策所需的关键信息。
研究设计与样本
- 纳入2015至2020年发表、每组≥100例且至少一个主要结局来自新生儿核心结局集的试验。
- 由Cochrane Neonatal招募的评分者使用CONSORT 2010和CONSORT-Outcomes 2022相关条目评估报告完整性,每项试验由3名评分者评估。
主要发现
- 共完成36项试验的结局报告评估,由39名评分者完成。
- 报告完整性水平高度可变。
- 所有试验完整报告了主要结局测量领域、比较治疗组所用的统计方法和受试者流程。
- 仅28%试验完整报告最小重要差异,24%报告结局数据缺失,66%报告结局评估者盲法,42%报告结局多重性处理。
关键数值与推荐
- 纳入标准:新生儿试验,每组≥100例,2015至2020年发表,至少1个主要结局来自新生儿核心结局集。
- 报告完整性评估使用CONSORT 2010和CONSORT-Outcomes 2022中关于主要结局定义、选择、测量、分析和解释的相关条目。
临床怎么用
- 试验者、期刊和同行评议者应使用现有结局报告指南以增强新生儿试验的透明报告。
- 临床解读新生儿试验时应关注最小重要差异、数据缺失、盲法和多重性处理等报告是否完整。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: Clear outcome reporting in clinical trials facilitates accurate interpretation and application of findings and improves evidence-informed decision-making. Standardized core outcomes for reporting neonatal trials have been developed, but little is known about how primary outcomes are reported in neonatal trials. Our aim was to identify strengths and weaknesses of primary outcome reporting in recent neonatal trials. METHODS: Neonatal trials including ≥100 participants/arm published between 2015 and 2020 with at least 1 primary outcome from a neonatal core outcome set were eligible. Raters recruited from Cochrane Neonatal were trained to evaluate the trials' primary outcome reporting completeness using relevant items from Consolidated Standards of Reporting Trials 2010 and Consolidated Standards of Reporting Trials-Outcomes 2022 pertaining to the reporting of the definition, selection, measurement, analysis, and interpretation of primary trial outcomes. All trial reports were assessed by 3 raters. Assessments and discrepancies between raters were analyzed. RESULTS: Outcome-reporting evaluations were completed for 36 included neonatal trials by 39 raters. Levels of outcome reporting completeness were highly variable. All trials fully reported the primary outcome measurement domain, statistical methods used to compare treatment groups, and participant flow. Yet, only 28% of trials fully reported on minimal important difference, 24% on outcome data missingness, 66% on blinding of the outcome assessor, and 42% on handling of outcome multiplicity. CONCLUSIONS: Primary outcome reporting in neonatal trials often lacks key information needed for interpretability of results, knowledge synthesis, and evidence-informed decision-making in neonatology. Use of existing outcome-reporting guidelines by trialists, journals, and peer reviewers will enhance transparent reporting of neonatal trials.
引用
- Baba A, Webbe J, Butcher NJ, et al. Heterogeneity and Gaps in Reporting Primary Outcomes From Neonatal Trials. Pediatrics. 2023. PMID: 37641881. doi:10.1542/peds.2022-060751
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。