Predicting frequent emergency department visits among children with asthma using EHR data
Predicting frequent emergency department visits among children with asthma using EHR data
复制标题
DOI：
10.1002/ppul.23735
复制
发表时间：
2017-07-01
影响因子：
3.1
通讯作者：
Grinspan, Zachary M.
中科院分区：
文献类型：
--
作者：
Das, Lala T.;Abramson, Erika L.;Grinspan, Zachary M.
目的对于患有哮喘的儿童来说，急诊室 (ED) 就诊很常见，而且费用昂贵，而且通常是可以避免的。尽管有几个因素与急诊室使用相关（人口统计、合并症、保险、药物），但使用电子健康记录 (EHR) 数据的预测性尚未得到充分研究。方法我们使用回顾性队列研究设计和来自一个中心的 EHR 数据，使用双变量和多变量统计来检查 1 年（2013 年）患者因素与下一年（2014 年）频繁使用急诊室（2 次就诊）的可能性之间的关系。我们应用并比较了几种机器学习算法来预测 ED 的频繁使用，然后根据准确性、简约性和可解释性选择了一个模型。结果我们识别了 2691 名儿童。在双变量分析中，未来频繁使用急诊科与人口统计、合并症、保险状况、用药史和医疗资源的使用有关。机器学习算法具有非常好的 AUC（曲线下面积）值 [0.66-0.87]，但 PPV（阳性预测值）[48-70%] 较好，灵敏度较差 [16-27%]。我们最终的多变量逻辑回归模型包含两个变量：保险状况和之前使用 ED 的情况。对于公共保险患者，频繁使用急诊科的几率是私人保险患者的 3.1 [2.2-4.5] 倍。一年内急诊就诊 4 次以上的公共保险患者和一年内急诊就诊 6 次以上的私人保险患者，下一年频繁使用急诊的概率为 50%。该模型的 AUC 为 0.86，PPV 为 56%，敏感性为 23%。 结论 在患有哮喘的儿童中，之前频繁使用 ED 和保险状况强烈预测未来 ED 使用。
ObjectiveFor children with asthma, emergency department (ED) visits are common, expensive, and often avoidable. Though several factors are associated with ED use (demographics, comorbidities, insurance, medications), its predictability using electronic health record (EHR) data is understudied.MethodsWe used a retrospective cohort study design and EHR data from one center to examine the relationship of patient factors in 1 year (2013) and the likelihood of frequent ED use (2 visits) in the following year (2014), using bivariate and multivariable statistics. We applied and compared several machine-learning algorithms to predict frequent ED use, then selected a model based on accuracy, parsimony, and interpretability.ResultsWe identified 2691 children. In bivariate analyses, future frequent ED use was associated with demographics, co-morbidities, insurance status, medication history, and use of healthcare resources. Machine learning algorithms had very good AUC (area under the curve) values [0.66-0.87], though fair PPV (positive predictive value) [48-70%] and poor sensitivity [16-27%]. Our final multivariable logistic regression model contained two variables: insurance status and prior ED use. For publicly insured patients, the odds of frequent ED use were 3.1 [2.2-4.5] times that of privately insured patients. Publicly insured patients with 4+ ED visits and privately insured patients with 6+ ED visits in a year had 50% probability of frequent ED use the following year. The model had an AUC of 0.86, PPV of 56%, and sensitivity of 23%.ConclusionAmong children with asthma, prior frequent ED use and insurance status strongly predict future ED use.