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

Abstract

Background

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.

Objectives

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.

Methods

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.

Results

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.

Conclusion

Our study findings provide validation for a novel model's use in the prediction of in-hospital malnutrition.

Keywords

malnutrition - screening - artificial intelligence - hospital - validation

Protection of Human and Animal Subjects

This retrospective study was approved by the Cedars-Sinai Medical Center Institutional Review Board.

Note

Virtual Studio Code integrated development environment with Github Copilot, and Cursor integrated development environment with default code completion copilot were used when writing code. Coding assistance for the production of the supplemental decile figures was provided by Claude 3.7 Sonnet.

Publication History

Received: 19 January 2025

Accepted: 06 June 2025

Accepted Manuscript online:
16 June 2025

Article published online:
14 November 2025

• References

• 1 Pratt KJ, Hernandez B, Blancato R, Blankenship J, Mitchell K. Impact of an interdisciplinary malnutrition quality improvement project at a large metropolitan hospital. BMJ Open Qual 2020; 9 (01) e000735
• 2 Saunders J, Smith T. Malnutrition: causes and consequences. Clin Med (Lond) 2010; 10 (06) 624-627
• 3 Mueller C, Compher C, Ellen DM. American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) Board of Directors. A.S.P.E.N. clinical guidelines: Nutrition screening, assessment, and intervention in adults. JPEN J Parenter Enteral Nutr 2011; 35 (01) 16-24
• 4 Weiss AJ (Truven Health Analytics), Fingar KR (Truven Health Analytics), Barrett ML (M.L. Barrett, Inc.), Elixhauser A (AHRQ), Steiner CA (AHRQ), Guenter P (American Society for Parenteral and Enteral Nutrition), Brown MH (Baxter International, Inc.). Characteristics of Hospital Stays Involving Malnutrition, 2013. HCUP Statistical Brief #210. September 2016. Agency for Healthcare Research and Quality, Rockville, MD. http://www.hcup-us.ahrq.gov/reports/statbriefs/sb210-Malnutrition-Hospital-Stays-2013.pdf
• 5 Barker LA, Gout BS, Crowe TC. Hospital malnutrition: prevalence, identification and impact on patients and the healthcare system. Int J Environ Res Public Health 2011; 8 (02) 514-527
• 6 Henriksen C, Gjelstad IM, Nilssen H, Blomhoff R. A low proportion of malnourished patients receive nutrition treatment - results from nutritionDay. Food Nutr Res 2017; 61 (01) 1391667
• 7 Smith PE, Smith AE. High-quality nutritional interventions reduce costs. Healthc Financ Manage 1997; 51 (08) 66-69
• 8 Starke J, Schneider H, Alteheld B, Stehle P, Meier R. Short-term individual nutritional care as part of routine clinical setting improves outcome and quality of life in malnourished medical patients. Clin Nutr 2011; 30 (02) 194-201
• 9 Tobert CM, Mott SL, Nepple KG. Malnutrition diagnosis during adult inpatient hospitalizations: Analysis of a multi-institutional collaborative database of Academic Medical Centers. J Acad Nutr Diet 2018; 118 (01) 125-131
• 10 Lim SL, Ng SC, Lye J, Loke WC, Ferguson M, Daniels L. Improving the performance of nutrition screening through a series of quality improvement initiatives. Jt Comm J Qual Patient Saf 2014; 40 (04) 178-186
• 11 Cortés-Aguilar R, Malih N, Abbate M, Fresneda S, Yañez A, Bennasar-Veny M. Validity of nutrition screening tools for risk of malnutrition among hospitalized adult patients: A systematic review and meta-analysis. Clin Nutr 2024; 43 (05) 1094-1116
• 12 Uhl S, Siddique SM, McKeever L. et al. Malnutrition in Hospitalized Adults: A Systematic Review. Comparative Effectiveness Review No. 249. (Prepared by the ECRI–Penn Medicine Evidence-based Practice Center under Contract No. 75Q80120D00002.) AHRQ Publication No. 21(22)-EHC035. Rockville, MD: Agency for Healthcare Research and Quality; ; October 2021. Accessed June 18, 2025 at: https://doi.org/10.23970/AHRQEPCCER249
• 13 Academy of Nutrition and Dietetics. About the Academy of Nutrition and Dietetics. Accessed April 11, 2025 at: https://www.eatright.org/about-the-academy
• 14 Skipper A, Coltman A, Tomesko J. et al. Reprint of: Position of the Academy of Nutrition and Dietetics: Malnutrition (Undernutrition) Screening Tools for All Adults. J Acad Nutr Diet 2022; 122 (10S): S50-S54
• 15 Merker M, Felder M, Gueissaz L. et al. Association of baseline inflammation with effectiveness of nutritional support among patients with disease-related malnutrition: A secondary analysis of a randomized clinical trial. JAMA Netw Open 2020; 3 (03) e200663
• 16 Bharadwaj S, Ginoya S, Tandon P. et al. Malnutrition: laboratory markers vs nutritional assessment. Gastroenterol Rep (Oxf) 2016; 4 (04) 272-280
• 17 Janssen SM, Bouzembrak Y, Tekinerdogan B. Artificial intelligence in malnutrition: A systematic literature review. Adv Nutr 2024; 15 (09) 100264
• 18 Ferguson M, Capra S, Bauer J, Banks M. Development of a valid and reliable malnutrition screening tool for adult acute hospital patients. Nutrition 1999; 15 (06) 458-464
• 19 Cedars-Sinai. Facts and Figures. Accessed April 6, 2025 at: https://www.cedars-sinai.org/newsroom/facts-and-figures/
• 20 Bujang MA, Adnan TH. Requirements for minimum sample size for sensitivity and specificity analysis. J Clin Diagn Res 2016; 10 (10) YE01-YE06
• 21 Jimenez EY, Lamers-Johnson E, Long JM. et al. Predictive validity of the Academy of Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition indicators to diagnose malnutrition tool in hospitalized adults: a cohort study. Am J Clin Nutr 2024; 119 (03) 779-787
• 22 White JV, Guenter P, Jensen G, Malone A, Schofield M. Academy of Nutrition and Dietetics Malnutrition Work Group, A.S.P.E.N. Malnutrition Task Force, A.S.P.E.N. Board of Directors. Consensus statement of the Academy of Nutrition and Dietetics/American Society for Parenteral and Enteral Nutrition: characteristics recommended for the identification and documentation of adult malnutrition (undernutrition). J Acad Nutr Diet 2012; 112 (05) 730-738
• 23 Phillips W. Coding for Malnutrition in the Adult Patient: What the Physician Needs to Know. Practical Gastroenterology: Nutrition Issues in Gastroenterology, Series #133. 2014
• 24 Koçak B. Key concepts, common pitfalls, and best practices in artificial intelligence and machine learning: focus on radiomics. Diagn Interv Radiol 2022; 28 (05) 450-462
• 25 Anghel A, Papandreou N, Parnell T, Palma AD, Pozidis H. Benchmarking and optimization of gradient boosting decision tree algorithms. 2018 . arXiv:1809.04559. Accessed May 27, 2025 at: https://arxiv.org/abs/1809.04559
• 26 Ke G, Meng Q, Finley T. et al. LightGBM: A highly efficient gradient boosting decision tree. Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS '17). Curran Associates Inc., Red Hook, NY, USA, 3149-3157
• 27 Microsoft Corporation. LightGBM. Accessed May 27, 2025 at: https://lightgbm.readthedocs.io/en/latest/Advanced-Topics.html
• 28 Cederholm T, Jensen GL, Correia MITD. et al.; GLIM Core Leadership Committee, GLIM Working Group. GLIM criteria for the diagnosis of malnutrition - A consensus report from the global clinical nutrition community. J Cachexia Sarcopenia Muscle 2019; 10 (01) 207-217
• 29 Jensen GL, Cederholm T, Correia MITD. et al. GLIM consensus approach to diagnosis of malnutrition: A 5-year update. JPEN J Parenter Enteral Nutr 2025; 49 (04) 414-427
• 30 Zadak Z, Hyspler R, Ticha A, Vlcek J. Polypharmacy and malnutrition. Curr Opin Clin Nutr Metab Care 2013; 16 (01) 50-55
• 31 Kvamme JM, Grønli O, Florholmen J, Jacobsen BK. Risk of malnutrition is associated with mental health symptoms in community living elderly men and women: the Tromsø study. BMC Psychiatry 2011; 11: 112
• 32 Kevdzija M, Laviano A, Worf I, Schuh C, Tarantino S, Hiesmayr M. Indirect nutrition and mobility risks during hospitalization: An architectural perspective on the nutritionDay Study Findings. Nutrients 2023; 15 (06) 1527
• 33 Nakamura T, Itoh T, Yabe A. et al. Polypharmacy is associated with malnutrition and activities of daily living disability among daycare facility users: A cross-sectional study. Medicine (Baltimore) 2021; 100 (34) e27073
• 34 Wang G, Cheng S, Zhan X. OpenChat: Advancing open-source language models with mixed-quality data. arXiv preprint arXiv:2309.11235 (journal). 2023 . Accessed April 6, 2024 at: https://arxiv.org/abs/2309.11235
• 35 Huggingface. Openchat/Openchat-3.5-0106. Accessed Jun 25, 2025 at: https://huggingface.co/openchat/openchat-3.5-0106
• 36 Huggingface. Datasets/Biomistral/BioInstructQA. Accessed November 2, 2024 at: https://huggingface.co/datasets/BioMistral/BioInstructQA
• 37 Pang RY, Parrish A, Joshi N. et al. QuALITY: Question Answering with Long Input Texts, Yes! Cornell University Computer Science. arXiv:2112.08608. Accessed October 21, 2024 at: https://arxiv.org/abs/2112.08608
• 38 Ozenne B, Subtil F, Maucort-Boulch D. The precision–recall curve overcame the optimism of the receiver operating characteristic curve in rare diseases. J Clin Epidemiol 2015; 68 (08) 855-859
• 39 McDermott MBA, Zhang H, Hyldig Hansen L, Angelotti G, Gallifant J. “A Closer Look at AUROC and AUPRC under Class Imbalance”. 38th Conference on Neural Information Processing Systems (NeurIPS 2024). arXiv:2401.06091. arXiv e-prints
• 40 Pinker E. Reporting accuracy of rare event classifiers. NPJ Digit Med 2018; 1: 56
• 41 Efron B, Tibshirani RJ. An Introduction to the Bootstrap. Boca Raton, FL: Chapman & Hall/CRC Publishers; 1993
• 42 The SciPy Community. scipy.stats. Bootstrap. Accessed June 21, 2024 at: https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.bootstrap.html
• 43 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40 (05) 373-383
• 44 Davidson-Pilon C. Lifelines: survival analysis in Python. J Open Source Softw 2019; 4: 1317
• 45 Klein JP, Logan B, Harhoff M, Andersen PK. Analyzaing survival curves at a fixed point in time. Stat Med 2007; 26 (24) 4505-4519
• 46 Timsina P, Joshi HN, Cheng FY. et al. MUST-Plus: A machine learning classifier that improves malnutrition screening in acute care facilities. J Am Coll Nutr 2021; 40 (01) 3-12
• 47 Liou L, Scott E, Parchure P. et al. Assessing calibration and bias of a deployed machine learning malnutrition prediction model within a large healthcare system. NPJ Digit Med 2024; 7 (01) 149
• 48 Jin BT, Choi MH, Moyer MF, Kim DA. Predicting malnutrition from longitudinal patient trajectories with deep learning. PLoS ONE 2022; 17 (07) e0271487
• 49 Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol 2010; 5 (09) 1315-1316
• 50 Guenter P, Abdelhadi R, Anthony P. et al. Malnutrition diagnoses and associated outcomes in hospitalized patients: United States, 2018. Nutr Clin Pract 2021; 36 (05) 957-969
• 51 Marcason W. Should albumin and prealbumin be used as indicators for malnutrition?. J Acad Nutr Diet 2017; 117 (07) 1144
• 52 Alkhalaf M, Yu P, Yin M, Deng C. Applying generative AI with retrieval augmented generation to summarize and extract key clinical information from electronic health records. J Biomed Inform 2024; 156: 104662
• 53 Alkhalaf M, Zhang Z, Chang HR. et al. Malnutrition and its contributing factors for older people living in residential aged care facilities: Insights from natural language processing of aged care records. Technol Health Care 2023; 31 (06) 2267-2278
• 54 Chicco D, Jurman G. The ABC recommendations for validation of supervised machine learning results in biomedical sciences. Front Big Data 2022; 5: 979465
• 55 Schuetz P, Fehr R, Baechli V. et al. Individualised nutritional support in medical inpatients at nutritional risk: a randomised clinical trial. Lancet 2019; 393 (10188): 2312-2321
• 56 Valladares AF, Kilgore KM, Partridge J, Sulo S, Kerr KW, McCauley S. How a malnutrition quality improvement initiative furthers malnutrition measurement and care: Results from a hospital learning collaborative. JPEN J Parenter Enteral Nutr 2021; 45 (02) 366-371
• 57 Sulo S, Feldstein J, Partridge J, Schwander B, Sriram K, Summerfelt WT. Budget impact of a comprehensive nutrition-focused quality improvement program for malnourished hospitalized patients. Am Health Drug Benefits 2017; 10 (05) 262-270
• 58 Cerasale MT, Mansour A, Molitch-Hou E, Bernstein S, Nguyen T, Kao CK. Implementation of a real-time documentation assistance tool: Automated Diagnosis (AutoDx). Appl Clin Inform 2024; 15 (03) 501-510
• 59 Oliveira E, Hautala M, Henry J. et al. Enhancing referrals to the supplemental nutrition assistance program through clinical integration of a standards-based decision support system. Appl Clin Inform 2025; 16 (01) 167-176