Abstract
The aim was to study the factors related to substandard nutrition in critically ill patients and the application effect of enteral nutrition nursing model based on evidence-based medicine. A total of 137 cases of critically ill patients treated in our hospital from January 2023 to January 2024 were clinically selected as the research subjects, and they were divided into the standardized group and the non-standardized group based on the patients' enteral nutrition. We collected and compared the clinical data of patients in the 2 groups, analyzed the covariance of the difference indicators, included the indicators without covariance problems in the logistic regression model to analyze the factors related to the patients' enteral nutrition not meeting the standard, constructed the clinical prediction model, and presented it as a visualization with a column line diagram, and assessed the predictive ability of the column line diagram model through internal validation of the drawing of the subject characteristics curve (receiver operating characteristic curve). The evidence-based medicine enteral nutrition care model was implemented for the patients, and its clinical application effect was analyzed. There were significant differences (P < .05) in the comparison of Glasgow Coma score, modified Nutritional Risk in Critical Illness score, catecholamines, feeding intolerance, and Acute Physiology and Chronic Health Evaluation II scores between the 2 groups. All difference variables were analyzed for variance inflation factor covariance using the R language (R package: logreg6.2.0), and the variance inflation factor of each difference variable was ≤10 with tolerance ≥0.1, so there was no problem of covariance, and the above indexes could be included in the logistic regression model, and the results found that all of the above indexes were the independent influences on nutritional nonattainment in critically ill patients (P < .05). A column-line graph prediction model was established, and the area under the curve value in the receiver operating characteristic curve was 0.950 with a 95% CI of (0.910-0.989), thus indicating that this clinical prediction model has a good degree of risk prediction. The patients' various nutritional indexes and enteral nutrition tolerance rate after implementation were better than before implementation (P < .05). Glasgow Coma score, modified Nutritional Risk in Critical Illness score, catecholamines, feeding intolerance, and Acute Physiology and Chronic Health Evaluation II Score are all independent influencing factors affecting the occurrence of enteral nutrition substandard in critically ill patients. The above indicators can be used to screen high-risk groups for the clinic, based on the evidence-based medicine enteral nutrition nursing care model and can effectively improve the nutritional status of the patient and the degree of nutritional tolerance. It provides a theoretical basis for clinical practice.