Related Experiment Video
Updated: Jun 21, 2026
08:32
Adapting Human Videofluoroscopic Swallow Study Methods to Detect and Characterize Dysphagia in Murine Disease Models
Published on: March 1, 2015
21.1K
Development and Validation of a GULP-Based Predictive Model for Dehydration in Elderly Patients with Post-Stroke
Yan Liang1, Xin Dai1, Bing Wei1
1Speech and Language Rehabilitation Department, Beijing Rehabilitation Hospital Affiliated with Capital Medical University, Beijing, China.
British Journal of Hospital Medicine (London, England : 2005)
|January 25, 2025
Summary
This study developed a General Utility for Latent Process (GULP) model to predict dehydration in elderly stroke patients with dysphagia. The validated model accurately assesses dehydration risk, improving clinical intervention and patient outcomes.
Area of Science:
- Geriatric Medicine
- Neurology
- Biostatistics
Background:
- Elderly patients with post-stroke dysphagia (PSD) are at high risk of dehydration.
- Accurate dehydration risk assessment is crucial for timely clinical intervention and improved prognosis in PSD patients.
- General Utility for Latent Process (GULP) technology offers a novel approach for developing complex medical prediction models.
Purpose of the Study:
- To establish and validate a dehydration prediction model for elderly patients with PSD using GULP criteria.
- To enhance the accuracy of dehydration risk assessment in this vulnerable population.
- To provide a scientific basis for clinical decision-making and improve patient outcomes.
Main Methods:
- Retrospective analysis of 200 elderly patients with PSD.
- Patients were divided into modeling (n=160) and verification (n=40) groups.
- Logistic regression identified influencing factors (age, lesion location, muscle strength, homocysteine, swallowing score); GULP model construction; ROC curve analysis for validation.
Main Results:
- Key predictors identified: age, lesion location, muscle strength grade, homocysteine (Hcy), and swallowing function score.
- The GULP model demonstrated strong predictive performance: AUC of 0.934 (training set) and 0.867 (validation set).
- High sensitivity and specificity achieved in both sets, indicating reliable dehydration risk prediction.
Conclusions:
- A GULP-based dehydration prediction model was successfully established and validated for elderly patients with PSD.
- The model exhibits high accuracy and practical applicability in clinical settings.
- This tool can significantly aid in managing dehydration risk and improving the prognosis of post-stroke patients with dysphagia.
