Work

A logistic regression-based nomogram model incorporating clinical, dietary, and nutritional data for predicting postoperative prognosis in elderly patients of grade A tertiary hospital

Year: 2025

Type: article

Abstract: Objective To develop and validate a logistic regression model predicting postoperative malnutrition risk in elderly patients using clinical, dietary, and nutritional data. Methods We analyzed 241 elderly patients (lung cancer lobectomy/esophageal cancer resection) admitted from January 2024 to December 2024. Participants were randomized 7:3 into training ( n = 168) and validation ( n = 73) sets. Prognostic factors were identified via univariate analysis and multivariate logistic regression to build a predictive model. Performance was assessed using C-index, calibration curves, and receiver operating characteristic (ROC) analysis. Results Baseline characteristics were comparable between sets ( P > 0.05). Multivariate analysis identified number of daily food types, cereal intake, high-quality protein intake, body mass index, serum albumin, and pre-albumin as malnutrition predictors (all P < 0.05). The model achieved C-indices of 0.834 (training set) and 0.703 (validation set). The area under the ROC curves were 0.834 (95% CI: 0.760–0.908) and 0.703 (95% CI: 0.539–0.866), respectively, with good calibration curve fit. Conclusion This validated model effectively predicts postoperative malnutrition risk in elderly surgical patients. Its visualization tools simplify complex nutritional assessment, offering a practical solution for resource-limited settings to improve postoperative care in grade A tertiary hospitals.

Source: Frontiers in Nutrition

DOI: 10.3389/fnut.2025.1644583