Patient Stratification Using Electronic Health Records from a Chronic Disease Management Program.
Patient Stratification Using Electronic Health Records from a Chronic Disease Management Program.
复制标题
使用慢性病管理计划中的电子健康记录对患者进行分层。
DOI：
10.1109/jbhi.2016.2514264
复制
发表时间：
2016
影响因子：
7.7
通讯作者：
Malin,Bradley
中科院分区：
文献类型：
--
作者：
Chen,Robert;Sun,Jimeng;Dittus,RobertS;Fabbri,Daniel;Kirby,Jacqueline;Laffer,CherylL;McNaughton,CandaceD;Malin,Bradley
目的本研究的目标是设计一个机器学习框架，以协助护理协调项目进行预后分层，以设计和提供个性化护理计划，并有效分配财务和医疗资源。 材料和方法本研究基于范德比尔特大学医学中心慢性护理协调项目中 2,521 名高血压患者的去识别队列。患者被建模为六年期间电子健康记录 (EHR) 中衍生的特征向量。我们应用逐步回归来确定与项目注册后平均动脉压降低至少 2 mmHg 相关的风险因素。随后通过逻辑回归分类器验证所得特征。最后，通过基于模型的聚类应用风险因素对患者进行分组。结果我们确定了一组预测特征，其中包括人口统计、药物和诊断概念的组合。对这些特征进行逻辑回归得出 ROC 曲线下面积 (AUC) 为 0.71 (95% CI: [0.67, 0.76])。基于这些特征，通过聚类确定了四个有临床意义的组——其中两个代表疾病较严重的患者，而其余的代表疾病较轻的患者。 讨论高血压患者的血压控制状态和对治疗的反应可能表现出显着的变化。然而，这项工作表明，聚类分析可以产生更同质的患者群体，这可能有助于临床医生设计和实施定制的护理计划。结论该研究表明，使用 EHR 数据进行预测建模和聚类有助于为护理提供者提供系统的、通用的方法，以根据患者层面的因素定制其管理方法。
ObjectiveThe goal of this study is to devise a machine learning framework to assist care coordination programs in prognostic stratification to design and deliver personalized care plans and to allocate financial and medical resources effectively.Materials and MethodsThis study is based on a de-identified cohort of 2,521 hypertension patients from a chronic care coordination program at the Vanderbilt University Medical Center. Patients were modeled as vectors of features derived from electronic health records (EHRs) over a six-year period. We applied a stepwise regression to identify risk factors associated with a decrease in mean arterial pressure of at least 2 mmHg after program enrollment. The resulting features were subsequently validated via a logistic regression classifier. Finally, risk factors were applied to group the patients through model-based clustering.ResultsWe identified a set of predictive features that consisted of a mix of demographic, medication, and diagnostic concepts. Logistic regression over these features yielded an area under the ROC curve (AUC) of 0.71 (95% CI: [0.67, 0.76]). Based on these features, four clinically meaningful groups are identified through clustering - two of which represented patients with more severe disease profiles, while the remaining represented patients with mild disease profiles.DiscussionPatients with hypertension can exhibit significant variation in their blood pressure control status and responsiveness to therapy. Yet this work shows that a clustering analysis can generate more homogeneous patient groups, which may aid clinicians in designing and implementing customized care programs.ConclusionThe study shows that predictive modeling and clustering using EHR data can be beneficial for providing a systematic, generalized approach for care providers to tailor their management approach based upon patient-level factors.