| 基于现实世界研究中临床随访数据的两种动态预测建模方法的实证研究 |
| |
| 引用本文： | 杨锋,陈欣,尤东方,黄丽红,魏朝晖,赵杨.基于现实世界研究中临床随访数据的两种动态预测建模方法的实证研究[J].中国临床医学,2021,28(5):751-756. |
| |
| 作者姓名： | 杨锋 陈欣 尤东方 黄丽红 魏朝晖 赵杨 |
| |
| 作者单位： | 南京医科大学公共卫生学院生物统计学系, 南京 211166;嘉兴泰格数据管理有限公司, 嘉兴 314000;南京医科大学公共卫生学院生物统计学系, 南京 211166;南京医科大学医学大数据中心, 南京 210000;复旦大学附属中山医院生物统计室, 上海 200032;CSCO生物统计学专家委员会RWE项目组;杭州泰格医药科技股份有限公司, 杭州 310000;CSCO生物统计学专家委员会RWE项目组;南京医科大学公共卫生学院生物统计学系, 南京 211166;南京医科大学医学大数据中心, 南京 210000;CSCO生物统计学专家委员会RWE项目组 |
| |
| 基金项目： | 国家自然科学基金（82173620，81903407），江苏省高校优势学科建设工程. |
| |
| 摘 要： | 目的： 针对现实世界研究（real-world study，RWS）中常见的具有纵向测量属性的动态观察指标，探讨界标法和联合建模法2种动态预测方法的应用价值。方法： 基于358例某重症肺炎患者预后数据，分别采用界标法和联合建模法，基于R软件，对于第5天、第10天、第15天尚处于观察期的某重症肺炎患者，预测其未来的死亡风险。结果： 2种方法均能在各时间点预测个体未来发生结局事件的概率。第5天、第10天和第15天，利用界标法进行动态预测的AUC分别为81.64%、85.89%和82.15%；而联合建模法的AUC分别为81.11%、85.07%和72.09%。结论： 在针对动态历史数据的现实世界研究中，可采用动态预测模型分析法，从而获得更为丰富的信息。
|
| 关 键 词： | 纵向数据 动态预测 界标法 联合建模 |
| 收稿时间： | 2021/9/6 0:00:00 |
| 修稿时间： | 2021/10/13 0:00:00 |
An empirical study on the two dynamic risk prediction methods based on clinical follow-up data in real-world study |
| |
| YANG Feng,CHEN Xin,YOU Dong-fang,HUANG Li-hong,WEI Zhao-hui,ZHAO Yang.An empirical study on the two dynamic risk prediction methods based on clinical follow-up data in real-world study[J].Chinese Journal Of Clinical Medicine,2021,28(5):751-756. |
| |
| Authors: | YANG Feng CHEN Xin YOU Dong-fang HUANG Li-hong WEI Zhao-hui ZHAO Yang |
| |
| Institution: | School of Public Health School, Nanjing Medical University, Nanjing 211166, Jiangsu, China;Jiaxing Tigermed Data Management Co. Ltd, Jiaxing 314000, Zhejiang, China;School of Public Health School, Nanjing Medical University, Nanjing 211166, Jiangsu, China;Center of Biomedical Big Data, Nanjing Medical University, Nanjing 210000, Jiangsu, China;Department of Biostatistics, Zhongshan Hospital, Fudan University, Shanghai 200032, China;CSCO Biostatistics Expert Committee RWE Working Group;Hangzhou Tigermed Consulting Co. Ltd, Hangzhou 310000, Zhejiang, China;CSCO Biostatistics Expert Committee RWE Working Group; School of Public Health School, Nanjing Medical University, Nanjing 211166, Jiangsu, China;Center of Biomedical Big Data, Nanjing Medical University, Nanjing 210000, Jiangsu, China;CSCO Biostatistics Expert Committee RWE Working Group |
| |
| Abstract: | Objective: To explore the applications of landmarking method and joint modeling for dynamic risk prediction in the analysis of longitudinal dataset obtained from real-world study. Methods: Based on the longitudinal prognosis data of 358 pneumonia patients, we used landmarking method and joint modeling respectively by software R, to estimate the probability of survival for the pneumonia patients who are under observation on day 5, day 10, and day 15. Results: Both of the two methods can make dynamic risk predictions on the probability of outcome in the future at different time points. On day 5, day 10, and day 15, the AUCs of the prediction using landmarking methods were 81.64%, 85.89%, and 82.15%, respectively, while those obtained from the joint modeling were 81.11%, 85.07%, and 72.09%, respectively. Conclusions: In the real-world study for dynamic historical data, dynamic prediction model analysis can be used to obtain more information. |
| |
| Keywords: | longitudinal data dynamic risk prediction landmark analysis joint modeling |
|
| 点击此处可从《中国临床医学》浏览原始摘要信息 |
| 点击此处可从《中国临床医学》下载免费的PDF全文 |
|
