Development of Electronic Health Record-Based Prediction Models for 30-Day Readmission Risk Among Patients Hospitalized for Acute Myocardial Infarction.
Development of Electronic Health Record-Based Prediction Models for 30-Day Readmission Risk Among Patients Hospitalized for Acute Myocardial Infarction.
复制标题
DOI：
10.1001/jamanetworkopen.2020.35782
复制
发表时间：
2021-01-04
影响因子：
13.8
通讯作者：
Brown JR
中科院分区：
文献类型：
--
作者：
Matheny ME;Ricket I;Goodrich CA;Shah RU;Stabler ME;Perkins AM;Dorn C;Denton J;Bray BE;Gouripeddi R;Higgins J;Chapman WW;MacKenzie TA;Brown JR
电子健康记录中部署的机器学习能否用于改善急性心肌梗死患者的再入院风险评估？在这项队列研究中，针对 10 187 名急性心肌梗死患者住院后 30 天的再入院情况，检查了经过外部验证的机器学习风险模型，在开发现场注意到了良好的辨别性能，但最佳辨别并没有带来最佳校准。外部验证使辨别力和校准能力显着下降。这项研究的结果强调，当使用机器学习模型来预测急性心肌梗死后再入院时，稳健的校准评估是对歧视的必要补充；即使存在通用数据模型，跨站点数据可用性的挑战也会限制外部验证性能。在美国，每年有超过60万成年人会经历急性心肌梗死（AMI），高达20%的患者会在30天内再次住院。这项研究强调了在这些风险模型中考虑校准的必要性。使用标准化为通用数据模型的电子健康记录 (EHR) 派生数据集来比较多个机器学习风险预测模型。这是一项回顾性队列研究，针对 2007 年 1 月 1 日至 2016 年 12 月 31 日期间从范德比尔特大学医学中心出院、初步诊断为 AMI 且未从其他机构转院的所有住院患者建立了 30 天再入院的风险预测模型。该模型于2011年4月2日至2016年12月31日在达特茅斯-希区柯克医疗中心进行了外部验证。数据分析发生在2019年1月4日至2020年11月15日之间。需要住院的急性心肌梗塞。主要结果是三十天后再次入院。从行政法规、用药医嘱和实验室测试中总共考虑了 141 个候选变量。使用参数模型（弹性网、最小绝对收缩和选择算子以及岭回归）和非参数模型（随机森林和梯度提升）开发了多种风险预测模型。使用保留数据以及接收者操作特征曲线下面积 (AUROC)、校准百分比和校准曲线带来评估模型。范德比尔特大学医学中心的最终队列包括 6163 名独特患者，其中平均 (SD) 年龄为 67 (13) 岁，其中 4137 名患者为男性 (67.1%)，1019 名患者 (16.5%) 为黑人或其他种族，933 名患者 (15.1%) 在 30 天内再次住院。达特茅斯-希区柯克医疗中心的最终队列包括 4024 名独特患者，平均 (SD) 年龄为 68 (12) 岁； 2584 人（64.2%）为男性，412 人（10.2%）在 30 天内再次住院，大多数队列是非西班牙裔和白人。最终测试集 AUROC 性能参数模型在 0.686 到 0.695 之间，非参数模型在 0.686 到 0.704 之间。在验证队列中，参数模型的 AUROC 性能介于 0.558 至 0.655 之间，非参数模型的 AUROC 性能介于 0.606 至 0.608 之间。在这项研究中，开发了 5 个机器学习模型并进行了外部验证，用于预测 AMI 30 天再入院住院情况。这些模型可以使用常规收集的数据部署在 EHR 中。该队列研究使用电子健康记录数据比较了多个外部验证的机器学习模型，以预测因急性心肌梗死住院的患者 30 天的再入院情况。
Can machine learning deployed in electronic health records be used to improve readmission risk estimation for patients following acute myocardial infarction? In this cohort study examining externally validated machine learning risk models for 30-day readmission of 10 187 patients following hospitalization for acute myocardial infarction, good discrimination performance was noted at the development site, but the best discrimination did not result in the best calibration. External validation yielded significant declines in discrimination and calibration. The findings of this study highlight that robust calibration assessments are a necessary complement to discrimination when machine learning models are used to predict post–acute myocardial infarction readmission; challenges with data availability across sites, even in the presence of a common data model, limit external validation performance. In the US, more than 600 000 adults will experience an acute myocardial infarction (AMI) each year, and up to 20% of the patients will be rehospitalized within 30 days. This study highlights the need for consideration of calibration in these risk models. To compare multiple machine learning risk prediction models using an electronic health record (EHR)–derived data set standardized to a common data model. This was a retrospective cohort study that developed risk prediction models for 30-day readmission among all inpatients discharged from Vanderbilt University Medical Center between January 1, 2007, and December 31, 2016, with a primary diagnosis of AMI who were not transferred from another facility. The model was externally validated at Dartmouth-Hitchcock Medical Center from April 2, 2011, to December 31, 2016. Data analysis occurred between January 4, 2019, and November 15, 2020. Acute myocardial infarction that required hospital admission. The main outcome was thirty-day hospital readmission. A total of 141 candidate variables were considered from administrative codes, medication orders, and laboratory tests. Multiple risk prediction models were developed using parametric models (elastic net, least absolute shrinkage and selection operator, and ridge regression) and nonparametric models (random forest and gradient boosting). The models were assessed using holdout data with area under the receiver operating characteristic curve (AUROC), percentage of calibration, and calibration curve belts. The final Vanderbilt University Medical Center cohort included 6163 unique patients, among whom the mean (SD) age was 67 (13) years, 4137 were male (67.1%), 1019 (16.5%) were Black or other race, and 933 (15.1%) were rehospitalized within 30 days. The final Dartmouth-Hitchcock Medical Center cohort included 4024 unique patients, with mean (SD) age of 68 (12) years; 2584 (64.2%) were male, 412 (10.2%) were rehospitalized within 30 days, and most of the cohort were non-Hispanic and White. The final test set AUROC performance was between 0.686 to 0.695 for the parametric models and 0.686 to 0.704 for the nonparametric models. In the validation cohort, AUROC performance was between 0.558 to 0.655 for parametric models and 0.606 to 0.608 for nonparametric models. In this study, 5 machine learning models were developed and externally validated to predict 30-day readmission AMI hospitalization. These models can be deployed within an EHR using routinely collected data. This cohort study compares multiple externally validated machine learning models using electronic health record data to predict 30-day readmission among patients hospitalized for acute myocardial infarction.
登录
查看更多内容
影响因子：
2.8
作者：
McManus, David D.;Saczynski, Jane S.;Kiefe, Catarina I.
通讯作者：
Kiefe, Catarina I.
影响因子：
4.8
作者：
Au, Anita G.;McAlister, Finlay A.;van Walraven, Carl
通讯作者：
van Walraven, Carl
影响因子：
11.3
作者：
Hannan, Edward L.;Zhong, Ye;King, Spencer B., III
通讯作者：
King, Spencer B., III
影响因子：
9.3
作者：
Collins GS;Reitsma JB;Altman DG;Moons KG
通讯作者：
Moons KG
DOI：
10.1093/jamia/ocz127
发表时间：
2019-12-01
影响因子：
6.4
作者：
Davis, Sharon E.;Greevy, Robert A., Jr.;Matheny, Michael E.
通讯作者：
Matheny, Michael E.