Artificial Intelligence-Based Hospital Malnutrition Screening: Validation of a Novel Machine Learning Model

Despite its morbidity, mortality, and financial burden, in-hospital malnutrition remains underdiagnosed and undertreated. Artificial intelligence (AI) offers a promising clinical informatics solution for identifying malnutrition risk and one that can be coupled with clinician-delivered patient care.The objectives of the study were to evaluate an AI-based hospital malnutrition screening model in a large and diverse inpatient population and to compare it to the currently used clinician-delivered malnutrition screening tool.We studied the performance of a gradient-boosted decision tree model incorporating a large language model (LLM) for feature extraction using the electronic medical record data of 106,449 patients over 3.75 years.The model's area under the receiver operating curve was 0.92 (95% confidence interval [CI]: 0.91-0.92) on the first day of hospitalization and rose to 0.95 (95% CI: 0.95-0.96) using the maximum risk predicted for each patient throughout hospitalization, indexed against discharge-coded malnutrition. Similar results were observed when indexed against dietitian-recorded malnutrition. The model outperformed the nurse-administered, modified version of the Malnutrition Screening Tool (MST) that was used in practice. Patients identified by the model had higher likelihoods of readmission and death compared with patients identified by the nurse-administered screener.Our study findings provide validation for a novel model's use in the prediction of in-hospital malnutrition.