文献速递类风湿关节炎脊柱关节炎其他结缔组织病

真实世界数据在患者安全研究中的应用

The use of real-world data to address questions of patient safety.
PubMed 文献 · Rheumatology (Oxford) · 2020-01-01

本文综述了生物制剂药物警戒注册登记在评估患者安全方面的优势与局限,指出注册登记数据统计效能高、可直接比较且不良事件细节丰富,在明确抗TNF治疗感染与恶性肿瘤风险中发挥核心作用,但存在选择偏倚、通道偏倚、病例捕获不全、未测量混杂及无法推断因果等局限。

要点

  • 注册登记数据统计效能高且不良事件细节丰富,可弥补试验 duration 短和样本同质的不足
  • 抗TNF治疗的感染与恶性肿瘤风险主要经注册登记数据阐明
  • 注册登记存在选择偏倚、通道偏倚、病例捕获不全和未测量混杂等局限

核心结论

  • 生物制剂药物警戒注册登记在评估患者安全方面具有重要价值,但存在选择偏倚、通道偏倚、病例捕获不全、未测量混杂及无法推断因果等局限。

研究设计与样本

  • 本文为综述,总结生物制剂药物警戒注册登记在患者安全问题上的优势与弱点。

主要发现

  • 注册登记统计效能高、可直接比较,并提供试验数据无法达到的不良事件细节水平。
  • 注册登记在阐明抗TNF治疗感染与恶性肿瘤风险方面发挥核心作用。

临床怎么用

  • 临床可借助注册登记数据理解生物制剂治疗选择的风险,同时需认识其偏倚与因果推断局限。
  • 电子健康记录和全国审计等常规收集数据可作为真实世界资源的补充。

局限性

  • 注册登记存在选择偏倚、通道偏倚、病例捕获不全、未测量混杂及无法推断因果等局限。

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

Pharmacovigilance registries of biologics were established to evaluate the risk of adverse events that may be missed in trials due to shorter durations and homogeneous samples. This review will present the strengths and weaknesses of registry data in addressing patient safety issues. Since their inception, scope has broadened because registries represent a relatively inexpensive approach to answering many clinical questions, both research and non-research focused. They achieve high statistical power, allow direct comparability, and offer a level of detail about adverse events not possible with trial data. Registries have been central in clarifying the risk of infection and malignancy with anti-TNF therapy, despite the limitations of selection and channelling bias, incomplete case capture, unmeasured confounding, and the inability to infer causality. Routinely collected data from electronic health records and national audits offer alternative real-world resources, further assisting patients and clinicians in understanding the risks of biologic therapy choices.

来源:Rheumatology (Oxford),查看原文。版权归原作者所有。

引用

  1. Yates M, Bechman K, Galloway J, et al. The use of real-world data to address questions of patient safety. Rheumatology (Oxford). 2020. PMID: 31834407. doi:10.1093/rheumatology/kez158

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。