Simulating complex patient populations with hierarchical learning effects to support methods development for post-market surveillance.
Simulating complex patient populations with hierarchical learning effects to support methods development for post-market surveillance.
复制标题
DOI：
10.1186/s12874-023-01913-9
复制
发表时间：
2023-04-11
影响因子：
4
通讯作者：
中科院分区：
文献类型：
--
作者：
验证新算法，例如将内在治疗风险与新治疗的经验学习相关的风险分开的方法，通常需要了解正在调查的数据特征的基本事实。由于在真实的世界数据中无法获得地面实况，因此使用模拟复杂临床环境的合成数据集进行模拟研究至关重要。我们描述和评估了一个可推广的框架内注入层次学习效应的强大的数据生成过程，结合了内在风险的大小，并占临床数据关系中的已知关键要素。我们提出了一个多步骤的数据生成过程，可定制的选项和灵活的模块，以支持各种仿真要求。具有非线性和相关特征的合成患者被分配到提供者和机构病例系列。根据用户定义，治疗概率和结局分配与患者特征相关。当引入新的治疗方法时，由于提供者和/或机构的经验学习而产生的风险以不同的速度和程度注入。为了进一步反映现实世界的复杂性，用户可以请求缺失值和省略变量。我们说明了我们的方法在一个案例研究中使用MIMIC-III数据的参考患者特征分布的实现。模拟数据中实现的数据特征反映了特定的值。治疗效果和特征分布的明显偏差，尽管不具有统计学显著性，但在小数据集（n < 3000）中最常见，可归因于小样本中估计实现值的随机噪声和变异性。当指定学习效应时，合成数据集显示不良结局概率的变化，因为受学习影响的治疗组病例增加，而不受学习影响的治疗组病例增加的概率稳定。我们的框架将临床数据模拟技术扩展到生成患者特征之外，以纳入分层学习效应。这使得开发和严格测试算法所需的复杂模拟研究成为可能，这些算法旨在将治疗安全信号从经验学习的影响中分离出来。通过支持这些努力，这项工作可以帮助确定培训机会，避免对医学进步的不必要限制，并加快治疗的改善。在线版本包含补充材料，可通过10.1186/s12874-023-01913-9获得。
Validating new algorithms, such as methods to disentangle intrinsic treatment risk from risk associated with experiential learning of novel treatments, often requires knowing the ground truth for data characteristics under investigation. Since the ground truth is inaccessible in real world data, simulation studies using synthetic datasets that mimic complex clinical environments are essential. We describe and evaluate a generalizable framework for injecting hierarchical learning effects within a robust data generation process that incorporates the magnitude of intrinsic risk and accounts for known critical elements in clinical data relationships. We present a multi-step data generating process with customizable options and flexible modules to support a variety of simulation requirements. Synthetic patients with nonlinear and correlated features are assigned to provider and institution case series. The probability of treatment and outcome assignment are associated with patient features based on user definitions. Risk due to experiential learning by providers and/or institutions when novel treatments are introduced is injected at various speeds and magnitudes. To further reflect real-world complexity, users can request missing values and omitted variables. We illustrate an implementation of our method in a case study using MIMIC-III data for reference patient feature distributions. Realized data characteristics in the simulated data reflected specified values. Apparent deviations in treatment effects and feature distributions, though not statistically significant, were most common in small datasets (n < 3000) and attributable to random noise and variability in estimating realized values in small samples. When learning effects were specified, synthetic datasets exhibited changes in the probability of an adverse outcomes as cases accrued for the treatment group impacted by learning and stable probabilities as cases accrued for the treatment group not affected by learning. Our framework extends clinical data simulation techniques beyond generation of patient features to incorporate hierarchical learning effects. This enables the complex simulation studies required to develop and rigorously test algorithms developed to disentangle treatment safety signals from the effects of experiential learning. By supporting such efforts, this work can help identify training opportunities, avoid unwarranted restriction of access to medical advances, and hasten treatment improvements. The online version contains supplementary material available at 10.1186/s12874-023-01913-9.
登录
查看更多内容
DOI：
10.1191/1740774504cn042oa
发表时间：
2004-01-01
期刊：
Clinical trials (London, England)
影响因子：
--
作者：
Cook, Jonathan A;Ramsay, Craig R;Fayers, Peter
通讯作者：
Fayers, Peter
影响因子：
24
作者：
Carroll, John D.;Edwards, Fred H.;Mack, Michael J.
通讯作者：
Mack, Michael J.
影响因子：
3.2
作者：
Benaim, Anat Reiner;Almog, Ronit;Beyar, Rafael
通讯作者：
Beyar, Rafael
影响因子：
24
作者：
Mao, Jialin;Redberg, Rita F.;Sedrakyan, Art
通讯作者：
Sedrakyan, Art
DOI：
10.1093/jamia/ocy142
发表时间：
2019-03-01
影响因子：
6.4
作者：
Baowaly, Mrinal Kanti;Lin, Chia-Ching;Chen, Kuan-Ta
通讯作者：
Chen, Kuan-Ta