Liverpool Centre for Cardiovascular Science, University of Liverpool and Liverpool Heart & Chest Hospital, Liverpool, UK.
Anthem Inc., Indianapolis, Indiana, USA.
Eur J Clin Invest. 2022 Aug;52(8):e13777. doi: 10.1111/eci.13777. Epub 2022 Apr 5.
To date, incident and recurrent MI remains a major health issue worldwide, and efforts to improve risk prediction in population health studies are needed. This may help the scalability of prevention strategies and management in terms of healthcare cost savings and improved quality of care.
We studied a large-scale population of 4.3 million US patients from different socio-economic and geographical areas from three health plans (Commercial, Medicare, Medicaid). Individuals had medical/pharmacy benefits for at least 30 months (2 years for comorbid history and followed up for 6 months or more for clinical outcomes). Machine-learning (ML) algorithms included supervised (logistic regression, neural network) and unsupervised (decision tree, gradient boosting) methodologies. Model discriminant validity, calibration and clinical utility were performed separately on allocated test sample (1/3 of original data).
In the absence of MI in comorbid history, the overall incidence rates were 0.442 cases/100 person-years and in the presence of MI history, 0.652. ML algorithms showed that supervised formulations had incrementally higher discriminant validity than unsupervised techniques (e.g., for incident MI outcome in the absence of MI in comorbid history: logistic regression "LR" - c index 0.921, 95%CI 0.920-0.922; neural network "NN" - c index 0.914, 95%CI 0.913-0.915; gradient boosting "GB" - c index 0.902, 95%CI 0.900-0.904; decision tree "DT" - c index 0.500, 95%CI 0.495-0.505). Calibration and clinical utility showed good to excellent results.
ML algorithms can substantially improve the prediction of incident and recurrent MI particularly in terms of the non-linear formulation. This approach may help with improved risk prediction, allowing implementation of cardiovascular prevention strategies across diversified sub-populations with different clusters of complexity.
迄今为止，全球范围内的新发和复发性心肌梗死仍然是一个主要的健康问题，需要努力提高人群健康研究中的风险预测能力。这可能有助于预防策略的可扩展性，以及在节省医疗成本和提高护理质量方面的管理。
我们研究了来自三个健康计划（商业、医疗保险、医疗补助）的来自不同社会经济和地理区域的 430 万美国患者的大型人群。个人至少有 30 个月的医疗/药房福利（对于共病病史为 2 年，对于临床结果为 6 个月或更长时间）。机器学习（ML）算法包括有监督（逻辑回归、神经网络）和无监督（决策树、梯度提升）方法。在分配的测试样本（原始数据的 1/3）上分别进行模型判别有效性、校准和临床实用性。
在没有共病病史的情况下，总体发生率为 0.442 例/100 人年，在有心肌梗死病史的情况下为 0.652。ML 算法表明，有监督配方比无监督技术具有递增的更高判别有效性（例如，在没有共病病史的情况下发生新发心肌梗死的结果：逻辑回归“LR”-c 指数 0.921，95%CI 0.920-0.922；神经网络“NN”-c 指数 0.914，95%CI 0.913-0.915；梯度提升“GB”-c 指数 0.902，95%CI 0.900-0.904；决策树“DT”-c 指数 0.500，95%CI 0.495-0.505）。校准和临床实用性显示出良好到优秀的结果。
机器学习算法可以大大提高新发和复发性心肌梗死的预测能力，特别是在非线性配方方面。这种方法可以帮助提高风险预测能力，从而在具有不同复杂程度的不同亚人群中实施心血管预防策略。
