Timsina Prem, Joshi Himanshu N, Cheng Fu-Yuan, Kersch Ilana, Wilson Sara, Colgan Claudia, Freeman Robert, Reich David L, Mechanick Jeffrey, Mazumdar Madhu, Levin Matthew A, Kia Arash
Institute for Healthcare Delivery Science, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
J Am Coll Nutr. 2021 Jan;40(1):3-12. doi: 10.1080/07315724.2020.1774821. Epub 2020 Jul 23.
Malnutrition among hospital patients, a frequent, yet under-diagnosed problem is associated with adverse impact on patient outcome and health care costs. Development of highly accurate malnutrition screening tools is, therefore, essential for its timely detection, for providing nutritional care, and for addressing the concerns related to the suboptimal predictive value of the conventional screening tools, such as the Malnutrition Universal Screening Tool (MUST). We aimed to develop a machine learning (ML) based classifier (MUST-Plus) for more accurate prediction of malnutrition.
A retrospective cohort with inpatient data consisting of anthropometric, lab biochemistry, clinical data, and demographics from adult (≥ 18 years) admissions at a large tertiary health care system between January 2017 and July 2018 was used. The registered dietitian (RD) nutritional assessments were used as the gold standard outcome label. The cohort was randomly split (70:30) into training and test sets. A random forest model was trained using 10-fold cross-validation on training set, and its predictive performance on test set was compared to MUST.
In all, 13.3% of admissions were associated with malnutrition in the test cohort. MUST-Plus provided 73.07% (95% confidence interval [CI]: 69.61%-76.33%) sensitivity, 76.89% (95% CI: 75.64%-78.11%) specificity, and 83.5% (95% CI: 82.0%-85.0%) area under the receiver operating curve (AUC). Compared to classic MUST, MUST-Plus demonstrated 30% higher sensitivity, 6% higher specificity, and 17% increased AUC.
ML-based MUST-Plus provided superior performance in identifying malnutrition compared to the classic MUST. The tool can be used for improving the operational efficiency of RDs by timely referrals of high-risk patients.
医院患者中的营养不良是一个常见但诊断不足的问题，会对患者的治疗结果和医疗费用产生不利影响。因此，开发高度准确的营养不良筛查工具对于及时发现营养不良、提供营养护理以及解决与传统筛查工具（如营养不良通用筛查工具（MUST））预测价值欠佳相关的问题至关重要。我们旨在开发一种基于机器学习（ML）的分类器（MUST-Plus），以更准确地预测营养不良。
使用一个回顾性队列，其包含2017年1月至2018年7月期间在一个大型三级医疗保健系统中成年（≥18岁）住院患者的人体测量、实验室生化、临床数据和人口统计学数据。注册营养师（RD）的营养评估用作金标准结局标签。该队列被随机分为（70:30）训练集和测试集。使用随机森林模型在训练集上进行10折交叉验证，并将其在测试集上的预测性能与MUST进行比较。
在测试队列中，总计13.3%的住院患者存在营养不良。MUST-Plus的灵敏度为73.07%（95%置信区间[CI]：69.61%-76.33%），特异度为76.89%（95%CI：75.64%-78.11%），受试者工作特征曲线下面积（AUC）为83.5%（95%CI：82.0%-85.0%）。与经典的MUST相比，MUST-Plus的灵敏度高30%，特异度高6%，AUC增加17%。
与经典的MUST相比，基于ML的MUST-Plus在识别营养不良方面表现更优。该工具可通过及时转诊高危患者来提高注册营养师的工作效率。
