Zilveren Kruis Achmea, Zeist, The Netherlands.
Department of Cardiology, Thorax Centre, Erasmus MC, University Medical Centre Rotterdam, Rotterdam, The Netherlands.
BMC Med Inform Decis Mak. 2021 Nov 1;21(1):303. doi: 10.1186/s12911-021-01657-w.
Accurately predicting which patients with chronic heart failure (CHF) are particularly vulnerable for adverse outcomes is of crucial importance to support clinical decision making. The goal of the current study was to examine the predictive value on long term heart failure (HF) hospitalisation and all-cause mortality in CHF patients, by exploring and exploiting machine learning (ML) and traditional statistical techniques on a Dutch health insurance claims database.
Our study population consisted of 25,776 patients with a CHF diagnosis code between 2012 and 2014 and one year and three years follow-up HF hospitalisation (1446 and 3220 patients respectively) and all-cause mortality (2434 and 7882 patients respectively) were measured from 2015 to 2018. The area under the receiver operating characteristic (ROC) curve (AUC) was calculated after modelling the data using Logistic Regression, Random Forest, Elastic Net regression and Neural Networks.
AUC rates ranged from 0.710 to 0.732 for 1-year HF hospitalisation, 0.705-0.733 for 3-years HF hospitalisation, 0.765-0.787 for 1-year mortality and 0.764-0.791 for 3-years mortality. Elastic Net performed best for all endpoints. Differences between techniques were small and only statistically significant between Elastic Net and Logistic Regression compared with Random Forest for 3-years HF hospitalisation.
In this study based on a health insurance claims database we found clear predictive value for predicting long-term HF hospitalisation and mortality of CHF patients by using ML techniques compared to traditional statistics.
准确预测哪些慢性心力衰竭（CHF）患者特别容易出现不良结局对于支持临床决策至关重要。本研究的目的是通过探索和利用荷兰医疗保险索赔数据库中的机器学习（ML）和传统统计技术，检查其对 CHF 患者长期心力衰竭（HF）住院和全因死亡率的预测价值。
我们的研究人群由 25776 名 2012 年至 2014 年间患有 CHF 诊断代码且在 2015 年至 2018 年间有一年和三年随访 HF 住院（分别为 1446 和 3220 名患者）和全因死亡率（分别为 2434 和 7882 名患者）的患者组成。使用 Logistic Regression、Random Forest、Elastic Net Regression 和 Neural Networks 对数据进行建模后，计算受试者工作特征曲线下的面积（AUC）。
1 年 HF 住院的 AUC 率范围为 0.710 至 0.732，3 年 HF 住院的 AUC 率为 0.705-0.733，1 年死亡率的 AUC 率为 0.765-0.787，3 年死亡率的 AUC 率为 0.764-0.791。Elastic Net 在所有终点中表现最好。技术之间的差异很小，仅在与 Random Forest 相比时，Elastic Net 与 Logistic Regression 之间在 3 年 HF 住院方面具有统计学意义。
在这项基于医疗保险索赔数据库的研究中，我们发现与传统统计学相比，使用 ML 技术对 CHF 患者的长期 HF 住院和死亡率具有明显的预测价值。
