文献速递免疫规划卫生政策与健康促进

应对COVID-19疫苗有效性研究中时变因素的分析方法与实例:世界卫生组织会议报告

Analytical approaches and examples of addressing time-varying factors in COVID-19 vaccine effectiveness studies: Report from a meeting of the World Health Organization.
PubMed 文献 · Vaccine · 2025-08-13

世界卫生组织于2024年5月7日召开虚拟会议,讨论在COVID-19疫苗有效性(VE)研究中如何区分日历时间与接种后时间这两个时变因素。会议以检测阴性设计研究为例,提出理论框架并展示三例研究,强调若日历时间被评估为混杂因素则须加以校正,接种后时间应作为暴露内在属性按类别报告。会议形成若干建议,涉及日历时间校正、接种后时间分类报告、总VE及变异株分层报告。

要点

  • 若日历时间被评估为潜在混杂因素,应加以校正。
  • 应按接种后时间分类报告COVID-19 VE,含中位时间及四分位距。
  • 应报告整个研究期间的总VE,并尽可能按变异株和疫苗类型分层。

核心结论

  • 会议强调若日历时间被评估为混杂因素则须校正以确保VE估计有效,接种后时间是暴露内在属性而非混杂因素,可按接种后时间类别评估。
  • 涉及协变量和分类的研究需要足够样本量,以确保同时评估时变因素对VE估计和解释的影响时达到所需精度。

研究设计与样本

  • 世界卫生组织于2024年5月7日召开虚拟会议,使用检测阴性设计COVID-19 VE研究实例讨论时变因素处理方法。
  • 会议以假设数据展示理论框架,并呈现三项近期检测阴性设计研究的数据。

主要发现

  • 三项研究展示了日历时间和COVID-19疫苗接种后时间对2023/24年秋季单价mRNA XBB.1.5加强针VE估计的潜在影响,背景为XBB和JN.1奥密克戎亚变异株共同流行。
  • 会议识别出两个可能影响COVID-19 VE估计及解释的时变因素:日历时间和接种后时间。

关键数值与推荐

  • 建议1:若日历时间被评估为潜在混杂因素,应加以校正。
  • 建议2:按COVID-19疫苗接种后时间分类报告VE,包括接种后中位时间和四分位距。
  • 建议3:报告整个研究期间的总COVID-19 VE,可对项目管理者有用。
  • 建议4:尽可能按SARS-CoV-2变异株或亚变异株提供VE,并按接种后时间分类和按COVID-19疫苗类型报告。

临床怎么用

  • 开展COVID-19 VE研究时,若日历时间被评估为潜在混杂因素应加以校正。
  • 报告VE时应按接种后时间分类并给出中位时间和四分位距,同时报告整个研究期间的总VE。
  • 尽可能按SARS-CoV-2变异株或亚变异株、接种后时间类别和疫苗类型分层报告VE。

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

As the COVID-19 pandemic has evolved, interpreting COVID-19 vaccine effectiveness (VE) has become increasingly complex due to time-varying factors that need to be disentangled, such as waning of protection and new SARS-CoV-2 variant or sub-variant emergence. Given that policymakers and regulatory authorities use COVID-19 VE data to inform COVID-19 vaccine and immunization program recommendations, the World Health Organization convened a virtual meeting on May 7, 2024. At the meeting, approaches for addressing time-varying factors were discussed using examples of COVID-19 VE studies applying a test-negative design. The meeting highlighted two distinct time-varying factors potentially influencing COVID-19 VE estimates and their interpretation: calendar time and time since vaccination. During the meeting, a theoretical framework for addressing these two time-varying factors was presented using hypothetical data. In addition, data from three recent test-negative design studies were presented. These studies illustrated the potential influence of calendar time and time since COVID-19 vaccination on VE estimates for the autumn 2023/24 monovalent mRNA XBB.1.5 booster. They did so in the context of co-circulating SARS-CoV-2 Omicron sub-variants, XBB and JN.1. The presentations reinforced the importance of accounting for calendar time if assessed as a confounder to ensure valid COVID-19 VE estimation. The presentations also emphasized that time since COVID-19 vaccination is an inherent property of the exposure (i.e., vaccination), not a confounder, which can be evaluated by time since COVID-19 vaccination categories. As for all studies involving covariates and categorization, sufficient sample size is required to ensure desired precision in simultaneously assessing the influence of time-varying factors on COVID-19 VE estimation and interpretation. The meeting resulted in several recommendations: 1) Account for calendar time if assessed as a potential confounder; 2) Report COVID-19 VE in time since COVID-19 vaccination categories, including median time and interquartile ranges since COVID-19 vaccination; 3) Report overall COVID-19 VE for the entire study period, which can be useful to program managers; 4) Provide COVID-19 VE by SARS-CoV-2 variant or sub-variant, if possible in time since COVID-19 vaccination categories and by COVID-19 vaccine type.

来源:Vaccine,查看原文。版权归原作者所有。

引用

  1. Varma A, Andrews NJ, Carazo S, et al. Analytical approaches and examples of addressing time-varying factors in COVID-19 vaccine effectiveness studies: Report from a meeting of the World Health Organization. Vaccine. 2025. PMID: 40812018. doi:10.1016/j.vaccine.2025.127567

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。