Nomogram Model for Screening the Risk of Frailty in Elderly Atrial Fibrillation Patients: A cross-sectional study

Abstract

Abstract Background Frailty is common in atrial fibrillation(AF)patients, but its related risk factors need to be further investigated. Furthermore, a risk prediction model based on risk factors urgently needed to be established to remind risk among AF patients. Purpose This study aimed to explore the multiple risk factors of frailty in elderly patients with atrial fibrillation(AF) and then construct and validate a nomogram risk prediction model to remind frailty events in this population. Methods A total of 337 hospitalized patients over 60(average age: 69, 53.1% male)with AF in Tianjin Medical University General Hospital from November 2021 to August 2022 were recruited. Patients were assessed for frailty with the FRAIL scale and then assigned into the groups, robust, pre-frail, and frail. The Least absolute shrinkage and selection operator (LASSO) and the Ordinal regression were utilized to screen independent risk factors. Subsequently, gather the factors in a nomogram to predict the risk among the AF population. The concordance index (C-index) and calibration curves were utilized to evaluate the performance of the nomogram model. Results The prevalence of frail and pre-frail were 23.1% and 52.2% among AF patients, respectively. A total of six predictors for frailty were screened out containing age, gender, history of coronary heart disease, number of chronic diseases, sleep disruption, and mental health status. The C-index of internal and external validation for the nomogram model were 0.821(95%CI: 0.778–0.864; bias corrected C-index: 0.795) and 0.837(95%CI: 0.780–0.893; bias corrected C-index: 0.774), respectively, indicating its favourable discriminative ability. Both internally and externally validated calibration charts were highly consistent with the ideal curve, illustrating that the model had a good predictive ability. Conclusion Frailty was common in the AF population, and the nomogram model has a great discriminative and predictive probability which can identify frailty risk incidents in elderly AF patients.