文献速递流行病学与生物统计真实世界研究与卫生经济

真实世界数据因果推断研究的偏倚风险评估工具:快速范围综述

Risk-of-bias tools for causal inference studies using real-world data: a rapid scoping review.
PubMed 文献 · J Clin Epidemiol · 2026-08-30

该快速范围综述检索2015年以来文献,纳入11个面向因果问题的偏倚风险评估工具,分析其结构、方法学侧重及在四阶段研究生成与四核心偏倚域中的覆盖情况,发现工具间差异显著,影响评估解读与比较。

要点

  • 纳入11个面向因果问题的RoB工具,多主要针对原始研究设计。
  • 定性或分类判断比数值评分更常见。
  • 工具在研究设计阶段覆盖最多,结果呈现阶段覆盖最少。

核心结论

  • 现有RoB工具在结构、方法学侧重及偏倚操作化方式上差异显著,解读和比较不同工具的评估时需考虑这些差异。

研究设计与样本

  • 快速范围综述,纳入2015年起发表的记录,包括描述明确为RWE开发或适用于因果推断的工具的研究与文件。

主要发现

  • 共纳入11个面向因果问题的工具,多数主要针对原始研究设计。
  • 定性或分类判断比数值评分更常见。
  • 研究设计阶段覆盖最频繁,结果呈现阶段覆盖最少。
  • 数据质量阶段由十个工具涉及,但覆盖程度因工具而异。
  • 无工具集中于单一核心偏倚域,多数工具将条目分布于多个域,选择与混杂偏倚的相对侧重因工具而异。

临床怎么用

  • 研究人员和决策者选择工具或比较不同工具的评估时,应考虑工具间在评估内容和偏倚评价方式上的显著差异。

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: This review aimed to identify and characterize tools designed, adapted, or validated to assess risk of bias (RoB) in causal real-world evidence, with a view to informing their potential use in evidence-informed decision-making. STUDY DESIGN AND SETTING: We conducted a rapid scoping review of records published from 2015 onwards. Eligible records included studies and documents describing tools explicitly developed for RWE or tools applicable to causal inference. Tool characteristics were extracted, and their items/domains were mapped across four stages of study generation and four core bias domains. RESULTS: Eleven tools oriented toward causal questions were included, most of which were designed primarily for primary studies. Qualitative or categorical judgments were more common than numerical scoring. Coverage of the stages of study generation was uneven: study design was the most frequently addressed stage, whereas results presentation was the least represented. The data quality stage was addressed by ten tools, although its coverage varied across instruments. Across core bias domains, no tool was concentrated in a single domain; most distributed items across several domains, although the relative emphasis placed on selection and confounding bias varied across instruments. A substantial number of items/domains could not be classified under the predefined core bias domains. CONCLUSION: Available RoB tools differ substantially in their structure, methodological emphasis, and the way they operationalize bias across stages of study generation and core bias domains. These differences should be considered when interpreting and comparing assessments conducted with different tools. PLAIN LANGUAGE SUMMARY: RWE uses information collected during routine healthcare to study whether healthcare interventions cause particular health outcomes. However, these studies may have problems in their design, data, or analysis that can lead to biased or misleading conclusions. Researchers use RoB tools to identify these potential problems and judge how trustworthy study results are. We reviewed 11 available tools that can be used to assess RoB in RWE with causality purpose. The tools differed considerably in what they assessed and how they evaluated potential bias. These differences should be considered when researchers and decision makers choose a tool or compare assessments made using different tools.

来源:J Clin Epidemiol,查看原文。版权归原作者所有。

引用

  1. Nieto-Gutierrez W, Moler-Zapata S, Medina-Ramirez SA, et al. Risk-of-bias tools for causal inference studies using real-world data: a rapid scoping review. J Clin Epidemiol. 2026. PMID: 42669371. doi:10.1016/j.jclinepi.2026.112486

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。