Liu Jiazhen, Finkelstein Joseph
Stevens Institute of Technology.
Columbia University.
Stud Health Technol Inform. 2018;247:880-884.
The main objective of this project was to introduce approaches for comprehensive medication risk assessment in people with polypharmacy that simultaneously account for multiple drug and gene effects. To achieve this goal, we developed an integrated knowledge repository of actionable pharmacogenes and a scoring algorithm that was pilot-tested using a data set containing pharmacogenomic information of people with polypharmacy. Metabolic phenotyping using resulting database demonstrated recall of 83.6% and precision of 87.1%. The final scoring algorithm yielded medication risk scores that allowed distinguish frequently hospitalized older adults with polypharmacy and older adults with polypharmacy with low hospitalization rate (average scores respectively: 75.89±15.45 and 10.51±1.82, p<0.05). The initial prototype assessment demonstrated feasibility of our approach and identified steps for improving risk scoring algorithms. Pharmacogenomics-driven medication risk assessment in patient with polypharmacy has potential in identifying inadequate drug regimens and preventing adverse drug events.
本项目的主要目标是引入针对多重用药患者的全面药物风险评估方法，该方法要同时考虑多种药物和基因效应。为实现这一目标，我们开发了一个可操作的药物基因组综合知识库以及一种评分算法，并使用一个包含多重用药患者药物基因组信息的数据集进行了初步测试。利用所得数据库进行代谢表型分析，召回率为83.6%，精确率为87.1%。最终的评分算法得出了药物风险评分，能够区分多重用药且频繁住院的老年人和多重用药但住院率低的老年人（平均得分分别为：75.89±15.45和10.51±1.82，p<0.05）。初步的原型评估证明了我们方法的可行性，并确定了改进风险评分算法的步骤。在多重用药患者中，由药物基因组学驱动的药物风险评估在识别不适当的药物治疗方案和预防药物不良事件方面具有潜力。
