Validation of an Electronic Health Record-Based Suicide Risk Prediction Modeling Approach Across Multiple Health Care Systems
Validation of an Electronic Health Record-Based Suicide Risk Prediction Modeling Approach Across Multiple Health Care Systems
复制标题
DOI：
10.1001/jamanetworkopen.2020.1262
复制
发表时间：
2020-03-25
影响因子：
13.8
通讯作者：
Smoller, Jordan W.
中科院分区：
文献类型：
--
作者：
Barak-Corren, Yuval;Castro, Victor M.;Smoller, Jordan W.
基于电子健康记录训练机器学习算法的过程能否在独立的医疗保健系统中识别出自杀企图风险增加的个体？结果在这项预后研究中，使用监督学习方法应用于结构化的电子健康记录数据，从超过370万名患者在5个不同的美国医疗保健系统，模型检测到平均38%的自杀企图的情况下，90%的特异性平均提前2.1年。这些研究结果表明，一种利用全方位结构化电子健康记录数据的计算效率高的机器学习方法可能能够检测到自杀患者的自杀行为风险，并可能促进临床决策支持工具的开发，为降低风险的干预措施提供信息。重要性自杀是死亡率的主要原因，近年来自杀相关死亡人数不断增加。个性化风险预测的自动化方法在解决这一日益严重的公共卫生威胁方面具有巨大潜力。为了促进它们的采用，它们必须首先在不同的卫生保健环境中得到验证。目的评估一种风险预测方法的通用性和跨站点性能，该方法使用来自电子健康记录的现成结构化数据预测多个独立的美国医疗保健系统中的事件自杀企图。设计、设置和参与者对于这项预后研究，数据提取自纵向电子健康记录数据，包括国际疾病分类，第九次修订诊断，实验室检查结果，程序代码和药物治疗，来自5个独立的医疗保健系统，参与了可扩展的研究共享健康网络。在各个研究中心，可获得截至2018年的6至17年的数据。结果由国际疾病分类第九次修订版编码定义，反映事件自杀企图（根据专家临床医生病历审查，阳性预测值>0.70）。在5个系统中的每一个中使用朴素贝叶斯分类器训练模型。在每个站点的独立数据集中交叉验证模型，并计算性能指标。数据分析于2017年11月至2019年8月进行。主要结局和测量主要结局是自杀未遂，其定义采用国际疾病分类第九次修订版编码。在每个站点测量预测的准确性和及时性。结果在5个卫生保健系统中，在纳入分析的3 & x202 f;714 & x202 f;105例患者（2 & x202 f;130 & x202 f;454例女性[57.2%]）中，确定了39 & x202 f;162例（1.1%）。预测特征因研究中心而异，但正如预期的那样，最常见的预测因子反映了精神健康状况（例如边缘型人格障碍，比值比为8.1-12.9，双相情感障碍，比值比为0.9-9.1）和物质使用障碍（例如药物戒断综合征，比值比为7.0-12.9）。尽管地理位置、人口统计学特征和人群健康特征存在差异，但各研究中心的模型性能相似，曲线下面积范围为0.71（95% CI，0.70-0.72）至0.76（95% CI，0.75-0.77）。在90%的特异性下，这些模型平均提前2.1年检测到38%的病例。结论和相关性在5个不同的医疗保健系统中，一种利用全方位结构化电子健康记录数据的计算效率高的方法能够检测到自杀患者的自杀行为风险。这种方法可以促进临床决策支持工具的发展，告知风险降低interventions.This预后研究评估的风险预测方法的性能，使用电子健康记录的数据来预测跨多个美国医疗保健系统的事件自杀企图。
Question Can a process for training machine-learning algorithms based on electronic health records identify individuals at increased risk of suicide attempts across independent health care systems? Findings In this prognostic study, using a supervised learning approach applied to structured electronic health record data from more than 3.7 million patients across 5 diverse US health care systems, models detected a mean of 38% of cases of suicide attempt with 90% specificity a mean of 2.1 years in advance. Meaning These findings suggest that a computationally efficient machine-learning approach leveraging the full spectrum of structured electronic health record data may be able to detect the risk of suicidal behavior in unselected patients and may facilitate the development of clinical decision support tools that inform risk reduction interventions.Importance Suicide is a leading cause of mortality, with suicide-related deaths increasing in recent years. Automated methods for individualized risk prediction have great potential to address this growing public health threat. To facilitate their adoption, they must first be validated across diverse health care settings. Objective To evaluate the generalizability and cross-site performance of a risk prediction method using readily available structured data from electronic health records in predicting incident suicide attempts across multiple, independent, US health care systems. Design, Setting, and Participants For this prognostic study, data were extracted from longitudinal electronic health record data comprising International Classification of Diseases, Ninth Revision diagnoses, laboratory test results, procedures codes, and medications for more than 3.7 million patients from 5 independent health care systems participating in the Accessible Research Commons for Health network. Across sites, 6 to 17 years' worth of data were available, up to 2018. Outcomes were defined by International Classification of Diseases, Ninth Revision codes reflecting incident suicide attempts (with positive predictive value >0.70 according to expert clinician medical record review). Models were trained using naive Bayes classifiers in each of the 5 systems. Models were cross-validated in independent data sets at each site, and performance metrics were calculated. Data analysis was performed from November 2017 to August 2019. Main Outcomes and Measures The primary outcome was suicide attempt as defined by a previously validated case definition using International Classification of Diseases, Ninth Revision codes. The accuracy and timeliness of the prediction were measured at each site. Results Across the 5 health care systems, of the 3 & x202f;714 & x202f;105 patients (2 & x202f;130 & x202f;454 female [57.2%]) included in the analysis, 39 & x202f;162 cases (1.1%) were identified. Predictive features varied by site but, as expected, the most common predictors reflected mental health conditions (eg, borderline personality disorder, with odds ratios of 8.1-12.9, and bipolar disorder, with odds ratios of 0.9-9.1) and substance use disorders (eg, drug withdrawal syndrome, with odds ratios of 7.0-12.9). Despite variation in geographical location, demographic characteristics, and population health characteristics, model performance was similar across sites, with areas under the curve ranging from 0.71 (95% CI, 0.70-0.72) to 0.76 (95% CI, 0.75-0.77). Across sites, at a specificity of 90%, the models detected a mean of 38% of cases a mean of 2.1 years in advance. Conclusions and Relevance Across 5 diverse health care systems, a computationally efficient approach leveraging the full spectrum of structured electronic health record data was able to detect the risk of suicidal behavior in unselected patients. This approach could facilitate the development of clinical decision support tools that inform risk reduction interventions.This prognostic study evaluates the performance of a risk prediction method using data from electronic health records to predict incident suicide attempts across multiple US health care systems.