Abstract: Objective To investigate the risk of frailty in patients with spinal cord injury and analyze the associated risk factors. Methods A total of 121 patients with spinal cord injury who were admitted to Jiangsu Province Hospital from June 2023 to September 2024 were selected in the study. Frailty was assessed by the Frailty Index (FI). Based on the occurrence of frailty, the patients were categorized into frailty and non-frailty groups. Their clinical data were collected. Multivariate logistic regression analysis was used to identify the influencing factors of frailty in spinal cord injury patients. A prediction model was constructed, and a nomogram was drawn. The model calibration was evaluated using the Hosmer-Lemeshow test. Receiver operating characteristic (ROC) curves were plotted to assess the predictive value of the nomogram for frailty in spinal cord injury patients. Additionally, clinical data of 60 spinal cord injury patients from the same hospital were used as an external validation set. Results Among the 121 patients, 34 were classified as frail (frailty group), with a frailty incidence of 28.10%. The remaining 87 patients had no frailty (non-frailty group). Patients in the frailty group were older and had a higher body mass index (BMI) than those in the non-frailty group (P<0.05). There were no statistical differences in sex, education background, and other general information between the two groups (P>0.05). The frailty group had a longer disease course than the non-frailty group, with statistical differences between the two groups in terms of the American Spinal Injury Association (ASIA) classification and the extent of spinal cord injury (P<0.05). No significant differences were found between the two groups regarding injury site, cause of injury, and other clinical data (P>0.05). Multivariate logistic regression analysis showed that age and ASIA classification were influencing factors for frailty in spinal cord injury patients (P<0.05). Accordingly, a prediction model for frailty in spinal cord injury patients was constructed: P = 1/(1 + e^-Z), Z = 0.049 × age + 0.650 × ASIA classification - 12.777. The Hosmer-Lemeshow goodness-of-fit test showed χ² values of 15.413 for the training set and 3.600 for the validation set, with P-values of 0.052 and 0.891, respectively. The prediction curve and its corrected prediction version closely approximated the standard curve, demonstrating good model calibration. ROC curve analysis revealed that the area under the curve (AUC) for the nomogram in predicting frailty in spinal cord injury patients was 0.821 for the training set and 0.808 for the validation set. The sensitivity and specificity of 94.12% and 68.97% for the training set, and 76.47% and 76.74% for the validation set, respectively. Conclusions Frailty incidence is high in spinal cord injury patients. Age and ASIA classification are significant risk factors for frailty in spinal cord injury patients, and clinical attention should be given to these factors.