文献速递糖尿病

真实世界肥胖识别:机器学习算法的开发与内部验证

Real‐World Identification of Obesity: Development and Internal Validation of a Machine Learning Algorithm
RSS 订阅 · Diabetes, Obesity and Metabolism · 2026-10-26

本研究利用美国商业及Medicare Advantage医保理赔与电子健康记录数据,开发并内部验证机器学习模型以回顾性识别BMI≥30 kg/m²的肥胖人群。共纳入1 728 010人,其中肥胖者706 992人(40.9%)。折叠分类模型的F1评分为0.797,优于仅用BMI诊断编码的混淆矩阵方法(F1评分0.716)。

要点

  • 机器学习模型识别肥胖的F1评分为0.797,优于仅用BMI诊断编码的0.716。
  • 研究人群肥胖患病率为40.9%,其中1级肥胖占21.5%,2级占10.9%,3级占8.5%。
  • 该模型可用于回顾性识别肥胖人群,辅助评估肥胖相关负担与结局。

核心结论

  • 机器学习模型成功识别肥胖人群,且表现优于仅使用BMI诊断编码的混淆矩阵方法。
  • 该方法有望增强回顾性肥胖识别能力,并指导肥胖相关负担、医疗利用、费用及结局的人群层面评估。

研究设计与样本

  • 本研究使用美国Healthcare Integrated Research Database中的商业及Medicare Advantage理赔和电子健康记录数据,并整合社区调查与食物可及性数据。
  • 数据时间范围为2022年1月1日至2024年9月30日。
  • 采用极端梯度提升(XGBoost)决策树机器学习方法建立多个模型,随机选择30%个体用于测试,70%用于训练与验证。

主要发现

  • 共纳入1 728 010人,其中706 992人(40.9%)为肥胖人群。
  • 1级肥胖371 843人(21.5%),2级肥胖187 857人(10.9%),3级肥胖147 292人(8.5%)。
  • 折叠分类模型的F1评分为0.797,优于仅使用BMI诊断编码的混淆矩阵方法的F1评分0.716。

关键数值与推荐

  • 肥胖定义为体重指数(BMI)≥30 kg/m²。
  • 模型性能指标包括精确率-召回率曲线下面积、F1评分、敏感性、特异性、阳性预测值和准确率。

临床怎么用

  • 该机器学习模型可用于回顾性识别数据库中未记录肥胖诊断的肥胖个体。
  • 可用于人群层面评估肥胖相关疾病负担、医疗资源利用、费用及健康结局。

局限性

  • 本研究仅进行内部验证,缺乏外部验证数据。
  • 研究人群来自美国商业及Medicare Advantage数据,结果在其他人群中的适用性尚不明确。

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

ABSTRACT Aims To develop and internally validate machine learning (ML) models to retrospectively identify people with obesity (PwO) (body mass index [BMI] ≥ 30 kg/m2) using administrative claims and electronic health record (EHR) data. Materials and Methods This study used US commercial and Medicare Advantage claims and EHR data from the Healthcare Integrated Research Database (HIRD) enriched with clinical, laboratory and social drivers of health data from the American Community Survey and Food Access Research Atlas between 1 January 2022 and 30 September 2024. Potential predictor features were defined and calculated for the study population. Several models were created with a decision tree-based ML method, Extreme Gradient Boosting (XGBoost). A random selection of 30% of individuals and their observed BMIs were used for testing; the remaining 70% were utilized for training and validation. Performance metrics included area under the precision-recall curve, F1 score, sensitivity, specificity, positive predictive value (PPV) and accuracy. F1 score, which balances sensitivity and PPV, was used to compare ML models to a confusion matrix using BMI diagnosis codes alone. Results Among 1 728 010 included individuals, 706 992 (40.9%) were PwO (Class 1, 371 843 [21.5%]; Class 2, 187 857 [10.9%]; Class 3, 147 292 [8.5%]). The collapsed categorical model (F1 score, 0.797) performed better than a confusion matrix using BMI diagnosis codes alone (F1 score, 0.716). Conclusions The ML model successfully identified PwO and outperformed a confusion matrix approach using BMI diagnostic codes alone, showing potential to enhance retrospective obesity identification and guide population-level assessment of obesity-related burden, healthcare utilization, costs and outcomes.

来源:Diabetes, Obesity and Metabolism,查看原文。版权归原作者所有。

引用

  1. Diabetes, Obesity and Metabolism:Real‐World Identification of Obesity: Development and Internal Validation of a Machine Learning Algorithm

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。