Construction of an Amputation Risk Prediction Model for Patients with Diabetic Foot (DF) in a Civil Aviation Hospital

Objective:To construct a prediction model to predict the amputation risk of patients with diabetic foot (DF) in a civil aviation hospital, identify key influencing factors of amputation risk, and predict the probability of amputation, thereby providing a reference for condition evaluation and medical management of DF patients. Methods:Basic data, DF data, and laboratory index data were collected from 48 DF patients (selected based on inclusion and exclusion criteria) who visited a civil aviation hospital in Chaoyang District, Beijing, from 2023 to 2024. Descriptive statistics were performed to establish baseline characteristics, followed by univariate analysis (comparison between groups). Variables with P-values less than 0.05 were selected for multivariate binary logistic regression analysis to form a regression model. The prediction performance of the model was analyzed to establish a robust prediction model that provides a reference for predicting amputation risk in DF patients. Results:Whether the Wagner score is greater than 3, the presence of PAD, and abnormal CRP levels were identified as independent risk factors for amputation in DF patients. A prediction model was formed with an AUC of 0.970, indicating good performance. This model can serve as an effective tool for predicting amputation risk in DF patients and provide a reference for clinical decision-making.