Fronc Maciej, Jakubczyk Michał, Love Sharon B, Talbot Susan, Rolfe Timothy
Decision Analysis and Support Unit, Institute of Econometrics, SGH Warsaw School of Economics, Warsaw, Poland.
Central Monitoring and Data Analytics, GSK, Warsaw, Poland.
Clin Trials. 2025 Jun;22(3):342-351. doi: 10.1177/17407745241304059. Epub 2025 Jan 2.
Clinical trials handle a huge amount of data which can be used during the trial to improve the ongoing study conduct. It is suggested by regulators to implement the remote approach to evaluate clinical trials by analysing collected data. Central statistical monitoring helps to achieve that by employing quantitative methods, the results of which are a basis for decision-making on quality issues.
This article presents a scoping review which is based on a systematic and iterative approach to identify and synthesise literature on central statistical monitoring methodology. In particular, we investigated the decision-making processes (with emphasis on quality issues) of central statistical monitoring methodology and its place in the clinical trial workflow. We reviewed papers published over the last 10 years in two databases (Scopus and Web of Science) with a focus on data mining algorithms of central statistical monitoring and its benefit to the quality of trials.
As a result, 24 scientific papers were selected for this review, and they consider central statistical monitoring at two levels. First, the perspective of the central statistical monitoring process and its location in the study conduct in terms of quality issues. Second, central statistical monitoring methods categorised into practices applied in the industry, and innovative methods in development. The established methods are discussed through the prism of categories of their usage. In turn, the innovations refer to either research on new methods or extensions to existing ones.
Our review suggests directions for further research into central statistical monitoring methodology - including increased application of multivariate analysis and using advanced distance metrics - and guidance on how central statistical monitoring operates in response to regulators' requirements.
临床试验处理大量数据，这些数据可在试验期间用于改进正在进行的研究。监管机构建议采用远程方法，通过分析收集的数据来评估临床试验。中央统计监测有助于通过采用定量方法来实现这一目标，其结果是质量问题决策的基础。
本文提出了一项范围综述，该综述基于系统且迭代的方法来识别和综合关于中央统计监测方法的文献。特别是，我们研究了中央统计监测方法的决策过程（重点是质量问题）及其在临床试验工作流程中的位置。我们回顾了过去10年在两个数据库（Scopus和Web of Science）中发表的论文，重点关注中央统计监测的数据挖掘算法及其对试验质量的益处。
结果，本综述选择了24篇科学论文，它们从两个层面考虑中央统计监测。首先，从中央统计监测过程及其在研究开展中就质量问题而言的位置的角度。其次，中央统计监测方法分为行业应用的实践方法和正在开发的创新方法。既定方法通过其使用类别的视角进行讨论。反过来，创新则涉及新方法的研究或对现有方法的扩展。
我们的综述提出了中央统计监测方法进一步研究的方向——包括增加多变量分析的应用和使用先进的距离度量——以及关于中央统计监测如何根据监管机构要求运作的指导。
