Development and validation of frailty risk prediction model for elderly patients with coronary heart disease

Abstract

Abstract Objective To analyze the influential factors of frailty in elderly patients with coronary heart disease (CHD), develop a nomogram-based risk prediction model for this population, and validate its predictive performance. Methods A total of 592 elderly patients with CHD were conveniently selected and enrolled from 3 tertiary hospitals, 5 secondary hospitals, and 3 community health service centers in China between October 2022 and January 2023. Data collection involved the use of the general information questionnaire, the Frail scale, and the instrumental ability of daily living assessment scale. And the patients were categorized into two groups based on frailty, and χ2 test as well as logistic regression analysis were used to identify and determine the influencing factors of frailty. The nomograph prediction model for elderly patients with CHD was developed using R software (version 4.2.2). The Hosmer–Lemeshow test and the area under the receiver operating characteristic (ROC) curve were employed to assess the predictive performance of the model. Additionally, the Bootstrap resampling method was utilized to validate the model and generate the calibration curve of the prediction model. Results The prevalence of frailty in elderly patients with CHD was 30.07%. The multiple factor analysis revealed that poor health status (OR = 28.169)/general health status (OR = 18.120), age (OR = 1.046), social activities (OR = 0.673), impaired instrumental ability of daily living (OR = 2.384) were independent risk factors for frailty (all P < 0.05). The area under the ROC curve of the nomograph prediction model was 0.847 (95% CI: 0.809 ~ 0.878, P < 0.001), with a sensitivity of 0.801, and specificity of 0.793; the Hosmer- Lemeshow χ2 value was 12.646 (P = 0.125). The model validation results indicated that the C value of 0.839(95% CI: 0.802 ~ 0.879) and Brier score of 0.139, demonstrating good consistency between predicted and actual values. Conclusion The prevalence of frailty is high among elderly patients with CHD, and it is influenced by various factors such as health status, age, lack of social participation, and impaired ability of daily life. These factors have certain predictive value for identifying frailty early and intervention in elderly patients with CHD.

Keywords

• Elderly
• Predictive model
• Frailty
• Coronary heart disease
• Risk factors