Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1922
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-19T18:14:20.553481Z
status: published
lens: p_space
sector: health
headline: Machine learning-based prediction of hospital-acquired malnutrition at discharge using the GLIM criteria: A multicenter real-world study
dek: Objective This study aimed to investigate the incidence and clinical impact of hospital-acquired malnutrition (HAM) and develop a machine learning model for early risk prediction. Methods This secondary analysis included prospectively collected data from a multicenter observational study involving 34 tertiary hospitals in China. Adult patients hospitalized for 7 to 30 days were enrolled. Patients without malnutrition at admission were evaluated for HAM at discharge according to the Global Leadership Initiative o…
gain_title: (none)
problem_title: Among patients not malnourished on admission, about one in five developed malnutrition during hospitalization, and these patients had more complications, longer stays and higher costs.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: Among patients not malnourished on admission, about one in five developed malnutrition during hospitalization, and these patients had more complications, longer stays and higher costs.
problem_evidence: 731 (21.6%) developed HAM during hospitalization | higher healthcare costs (¥48 980 versus ¥20 640)
quick_read: Researchers analysed prospectively collected data from 4746 adults hospitalized for 7 to 30 days at 34 tertiary hospitals in China. Patients without malnutrition at admission were reassessed at discharge using the GLIM criteria, and machine learning models were built from clinical and nutritional variables available on admission.

Malnutrition acquired in hospital appears to be a frequent and costly problem even among patients who arrive well nourished. The authors suggest the model could flag high-risk patients early for nutritional intervention, but the supplied text does not report the model's accuracy or show that acting on its predictions improves outcomes.
limitation: The excerpt does not report the model's performance figures, and it does not show that prediction-guided nutritional intervention improves outcomes; the authors only say the model 'may enable' early identification.
tag: Evidence-backed problem
key_points: At admission, 1364 of 4746 patients (28.7%) already had malnutrition. | Of 3382 patients without baseline malnutrition, 731 (21.6%) developed hospital-acquired malnutrition. | Patients with HAM had more hospital-acquired complications (9.0% versus 2.9%). | HAM was linked to longer stays (14.93 versus 13.53 days) and higher costs (¥48 980 versus ¥20 640).
rundown: The study is a secondary analysis of a multicenter observational dataset. Only adults with stays of 7 to 30 days were enrolled, and HAM was defined using the GLIM criteria at discharge.

Model development used variables available at admission. Performance was assessed through discrimination, calibration and clinical utility analyses, though the supplied excerpt does not give the results of those analyses.

The comparison between patients with and without HAM covered complications, length of stay and healthcare costs. The authors conclude that HAM is common and that early identification could allow timely nutritional intervention.
sources:
- peer_reviewed | Nutrition | https://doi.org/10.1016/j.nut.2026.113413 | 2026-10-14
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

c09f2f42ef507ec1fcaf61e71f6bac274847602a093795cea7c40d55b1c7750b

previous

0000000000000000000000000000000000000000000000000000000000000000