Department of Industrial and Systems Engineering, University of Florida, Gainesville, Florida, USA.
Division of Cardiovascular Medicine, University of Florida, Gainesville, Florida, USA.
Clin Cardiol. 2021 Feb;44(2):230-237. doi: 10.1002/clc.23532. Epub 2020 Dec 23.
Readmission or death soon after heart failure (HF) admission is a significant problem. Traditional analyses for predicting such events often fail to consider the gamut of characteristics that may contribute- tending to focus on 30-day outcomes even though the window of increased vulnerability may last up to 90 days. Risk assessments incorporating machine learning (ML) methods may be better suited than traditional statistical analyses alone to sort through multitude of data in the electronic health record (EHR) and identify patients at higher risk. HYPOTHESIS: ML-based decision analysis may better identify patients at increased risk for 90-day acute HF readmission or death after incident HF admission. METHODS AND RESULTS: Among 3189 patients who underwent index HF hospitalization, 15.2% experienced primary or acute HF readmission and 11.5% died within 90 days. For risk assessment models, 98 variables were considered across nine data categories. ML techniques were used to help select variables for a final logistic regression (LR) model. The final model's AUC was 0.760 (95% CI 0.752 to 0.767), with sensitivity of 83%. This proved superior to an LR model alone [AUC 0.744 (95% CI 0.732 to 0.755)]. Eighteen variables were identified as risk factors including dilated inferior vena cava, elevated blood pressure, elevated BUN, reduced albumin, abnormal sodium or bicarbonate, and NT pro-BNP elevation. A risk prediction ML-based model developed from comprehensive characteristics within the EHR can efficiently identify patients at elevated risk of 90-day acute HF readmission or death for whom closer follow-up or further interventions may be considered.
心力衰竭（HF）入院后不久再次入院或死亡是一个重大问题。传统的预测此类事件的分析方法往往没有考虑到可能导致这种情况的各种特征，往往侧重于 30 天的结果，尽管易损期可能长达 90 天。纳入机器学习（ML）方法的风险评估可能比传统的统计分析更适合于梳理电子健康记录（EHR）中的大量数据并识别风险较高的患者。假设：基于 ML 的决策分析可能更好地识别出在 HF 入院后 90 天内发生急性 HF 再入院或死亡风险增加的患者。方法和结果：在 3189 名接受指数 HF 住院治疗的患者中，15.2%经历了原发性或急性 HF 再入院，11.5%在 90 天内死亡。对于风险评估模型，考虑了九个数据类别中的 98 个变量。使用 ML 技术来帮助选择最终逻辑回归（LR）模型的变量。最终模型的 AUC 为 0.760（95%CI 0.752 至 0.767），灵敏度为 83%。这优于单独的 LR 模型[AUC 0.744（95%CI 0.732 至 0.755）]。确定了 18 个风险因素，包括扩张的下腔静脉、血压升高、BUN 升高、白蛋白减少、钠或碳酸氢盐异常以及 NT pro-BNP 升高。从 EHR 中的综合特征开发的风险预测 ML 模型可以有效地识别出 90 天内发生急性 HF 再入院或死亡风险较高的患者，这些患者可能需要更密切的随访或进一步的干预。
