文献速递医学教育与科研方法

作者在分析研究报告报告质量时任意使用方法学方法:一项元研究

Authors arbitrarily used methodological approaches to analyze the quality of reporting in research reports: a meta-research study.
PubMed 文献 · J Clin Epidemiol · 2023-03-11

该元研究分析了截至2021年7月18日发表的356篇评估随机对照试验、系统综述和观察性研究报告质量的文献,发现88%聚焦特定主题领域,CONSORT清单最常用(67%),75%使用数值评分,11%使用不同阈值,47%分析依从性预测因素,方法学差异大,呼吁共识。

要点

  • 分析356篇评估报告质量的文献,方法学差异很大。
  • CONSORT清单最常用(225篇,67%),252篇(75%)使用数值评分。
  • 仅36篇(11%)使用报告质量阈值,158篇(47%)分析依从性预测因素。

核心结论

  • 用于评估证据报告质量的方法学差异很大,研究界需要就评估报告质量的一致方法学达成共识。

研究设计与样本

  • 分析了截至2021年7月18日发表的、使用PRISMA、CONSORT或STROBE清单评估证据报告质量的文献,共356篇。

主要发现

  • 356篇分析文章中,293篇(88%)调查了特定主题领域。
  • CONSORT清单最常用(N=225;67%),以原始、修改、部分形式或其扩展形式使用。
  • 252篇文章(75%)对清单条目依从性给予数值评分,其中36篇(11%)使用了不同的报告质量阈值。
  • 158篇(47%)文章分析了报告清单依从性的预测因素,最常研究的因素是文章发表年份(N=82;52%)。

临床怎么用

  • 原文未据此给出推荐。

局限性

  • 原文未报告局限性

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: Many authors used reporting checklists as an assessment tool to analyze the reporting quality of diverse types of evidence. We aimed to analyze methodological approaches used by researchers assessing reporting quality of evidence in randomized controlled trials, systematic reviews, and observational studies. STUDY DESIGN AND SETTING: We analyzed articles reporting quality assessment of evidence with Preferred Reporting Items of Systematic Reviews and Meta-Analyses (PRISMA), CONsolidated Standards of Reporting Trials (CONSORT), or the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) checklists published up to 18 July 2021. We analyzed methods used for assessing reporting quality. RESULTS: Among 356 analyzed articles, 293 (88%) investigated a specific thematic field. The CONSORT checklist (N = 225; 67%) was most often used, in its original, modified, partial form, or its extension. Numerical scores were given for adherence to checklist items in 252 articles (75%), of which 36 articles (11%) used various reporting quality thresholds. In 158 (47%) articles, predictors of adherence to reporting checklist were analyzed. The most studied factor associated with adherence to reporting checklist was the year of article publication (N = 82; 52%). CONCLUSION: The methodology used for assessing reporting quality of evidence varied considerably. The research community needs a consensus on a consistent methodology for assessing the quality of reporting.

来源:J Clin Epidemiol,查看原文。版权归原作者所有。

引用

  1. Plenkovic M, Civljak M, Puljak L, et al. Authors arbitrarily used methodological approaches to analyze the quality of reporting in research reports: a meta-research study. J Clin Epidemiol. 2023. PMID: 36907252. doi:10.1016/j.jclinepi.2023.03.008

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。