Analysis of risk factors and a predictive model of malnutrition in disabled stroke patients / 中华物理医学与康复杂志

ABSTRACT

Objective:To analyze the risk factors for malnutrition among disabled stroke survivors and devise a prediction model.Methods:A total of 373 disabled stroke survivors treated in the Department of Rehabilitation Medicine, the First Affiliated Hospital of Zhengzhou University in 2021 formed a control group ( n=102) and a malnutrition group ( n=271) according to their nutritional status. Univariate correlation analysis and multivariate logistic regression were used to analyze the risk factors for malnutrition and their predictive value. Results:Age, dysphagia, pulmonary infection, disability score and feeding style were found to be related significantly to the occurrence of malnutrition. Multivariate logistic regression confirmed that age, pulmonary infection, dysphagia, low total intake and a low Barthel index were useful predictors of malnutrition in such persons. Moreover, patients who had received nasogastric tube feeding were at much higher risk of malnutrition than those with intermittent oroesophageal tube feeding. The area under the receiver operating characteristics curve of the Barthel index combined with dysphagia to predict malnutrition was 0.84. The critical value was 0.67 with a sensitivity of 88% and a specificity of 72.5%.Conclusions:Age, pulmonary infection, dysphagia, feeding method, total intake and disability score are risk factors for malnutrition in disabled stroke survivors. The Barthel index combined with dysphagia has good predictive power for the occurrence of malnutrition in such persons.