# Development and validation of a frailty risk prediction model in patients with peripheral artery disease

## TL;DR

This study developed a model to predict frailty in patients with peripheral artery disease, helping clinicians identify high-risk individuals for targeted interventions.

## Contribution

A novel risk prediction model for frailty in peripheral artery disease patients was developed and validated.

## Key findings

- Age, hemoglobin level, comorbidities, and daily living activities were identified as independent risk factors for frailty.
- The model showed good discrimination and calibration in both training and validation sets.
- Decision curve analysis confirmed the model's clinical utility across various probability thresholds.

## Abstract

To investigate the current status and risk factors of frailty among patients with peripheral artery disease, and to develop a risk prediction model to inform targeted clinical interventions.

Patients were consecutively recruited for this investigation from August 2024 to May 2025. The study included 499 individuals with peripheral artery disease who were hospitalized in the vascular surgery department of a tertiary hospital in Southwest China. Data were collected using a general information questionnaire, laboratory test results, the Barthel Index, and the Social Support Rating Scale. The Tilburg Frailty Indicator was used to classify patients into a non-frailty group and a frailty group. The dataset was randomly split in a 7:3 ratio into a training set and a validation set. Independent predictors of frailty were identified through univariate and multivariate logistic regression analyses. The risk prediction model was developed using R software. Discrimination of the model was evaluated by plotting receiver operating characteristic (ROC) curves and calculating the area under the curve (AUC), sensitivity, and specificity in both the training and validation sets. Model calibration was assessed using the Hosmer–Lemeshow goodness-of-fit test and calibration curves. Clinical utility was evaluated using decision curve analysis.

Age, hemoglobin level, number of comorbidities, and activities of daily living were identified as independent risk factors for frailty. In the training set, the AUC was 0.771 (95% CI: 0.721–0.821), with a sensitivity of 0.788 and a specificity of 0.808. In the validation set, the AUC was 0.704 (95% CI: 0.620–0.788), with a sensitivity of 0.743 and a specificity of 0.682. The Hosmer–Lemeshow test indicated good calibration in both the training set (χ2 = 7.967, P = 0.435) and the validation set (χ2 = 9.642, P = 0.291). DCA showed that the model provided net clinical benefit within threshold probability ranges of 10%–80% in the training set and 20%–74% in the validation set.

The developed risk prediction model exhibited predictive performance and can assist clinical healthcare providers in identifying populations at high risk of frailty among patients with PAD, thereby providing a reference for developing intervention strategies targeting relevant risk factors.