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

## TL;DR

This study creates a model to predict malnutrition risk in elderly patients after surgery using clinical and dietary data, aiming to improve postoperative care.

## Contribution

A novel logistic regression-based nomogram model integrating clinical, dietary, and nutritional factors for postoperative malnutrition prediction in elderly patients.

## Key findings

- The model achieved C-indices of 0.834 in training and 0.703 in validation sets.
- Key predictors included food types, cereal intake, protein intake, BMI, and serum markers.
- Calibration curves showed good model fit and ROC analysis confirmed predictive accuracy.

## Abstract

To develop and validate a logistic regression model predicting postoperative malnutrition risk in elderly patients using clinical, dietary, and nutritional data.

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.

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.

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.