利用电子健康记录识别HIV暴露前预防候选者:超学习在罕见结局风险预测中的应用
研究基于Atrius Health 2007—2015年399 385名患者(150例HIV感染)的电子健康记录,提取180个协变量,用超学习等机器学习算法构建1年HIV感染风险预测模型。采用1:20类别比的lasso模型交叉验证AUC为0.86,优于超学习的0.84。
要点
- 基于电子健康记录常规变量可构建1年HIV感染风险预测模型
- 采用1:20类别比的lasso模型区分度最佳,cv-AUC达0.86
- 传统医师筛选23个主项的logistic回归模型cv-AUC为0.81,略逊
核心结论
- 机器学习可基于医师筛选的电子健康记录预测因子,成功构建1年HIV感染风险预测模型。
研究设计与样本
- 数据来自Atrius Health 2007—2015年399 385名患者,其中150例HIV感染,提取180个协变量。
- 训练42种复杂算法变体,采用不同抽样方案平衡病例对照比例,并以k折交叉验证比较超学习与各单一算法。
主要发现
- 采用1:20类别比的lasso模型交叉验证AUC为0.86,优于超学习的0.84。
- 仅含23个医师筛选主项的传统logistic回归模型交叉验证AUC为0.81。
- 超学习交叉验证AUC为0.84。
关键数值与推荐
- lasso(1:20类别比)cv-AUC=0.86;超学习cv-AUC=0.84;传统logistic回归(23个主项)cv-AUC=0.81。
临床怎么用
- 可基于电子健康记录常规采集信息辅助临床医生识别HIV暴露前预防候选者,但模型外推需谨慎。
局限性
- HIV感染结局极罕见(399 385人中150例),模型开发面临病例对照极度不平衡的统计挑战。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
UNLABELLED: Human immunodeficiency virus (HIV) pre-exposure prophylaxis (PrEP) protects high risk patients from becoming infected with HIV. Clinicians need help to identify candidates for PrEP based on information routinely collected in electronic health records (EHRs). The greatest statistical challenge in developing a risk prediction model is that acquisition is extremely rare. METHODS: Data consisted of 180 covariates (demographic, diagnoses, treatments, prescriptions) extracted from records on 399 385 patient (150 cases) seen at Atrius Health (2007-2015), a clinical network in Massachusetts. Super learner is an ensemble machine learning algorithm that uses k-fold cross validation to evaluate and combine predictions from a collection of algorithms. We trained 42 variants of sophisticated algorithms, using different sampling schemes that more evenly balanced the ratio of cases to controls. We compared super learner's cross validated area under the receiver operating curve (cv-AUC) with that of each individual algorithm. RESULTS: The least absolute shrinkage and selection operator (lasso) using a 1:20 class ratio outperformed the super learner (cv-AUC = 0.86 vs 0.84). A traditional logistic regression model restricted to 23 clinician-selected main terms was slightly inferior (cv-AUC = 0.81). CONCLUSION: Machine learning was successful at developing a model to predict 1-year risk of acquiring HIV based on a physician-curated set of predictors extracted from EHRs.
引用
- Gruber S, Krakower D, Menchaca JT, et al. Using electronic health records to identify candidates for human immunodeficiency virus pre-exposure prophylaxis: An application of super learning to risk prediction when the outcome is rare. Stat Med. 2020. PMID: 32578905. doi:10.1002/sim.8591
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。