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

Abstract Objective To analyze the influencing factors of frailty for elderly patients with coronary heart disease (CHD), development a risk prediction model of nomogram for elderly patients with CHD, and verify the prediction effect. Methods A total of 592 elderly patients with CHD from 3 third-level hospitals, 5 second-level hospitals and 3 community health service centers in Guizhou Province from April 2022 to January 2023 were conveniently selected and enrolled. The general information questionnaire, the Frail scale and the instrumental ability of daily living assessment scale were utilized to collect data. And the patients divided into two group according to frailty, χ² test and logistic regression analysis were used to select and determine the influencing factors of frailty. The nomograph prediction model of elderly patients with CHD was established by R software (4.2.2). Hosmer-Lemeshow test and the area under the ROC curve were used to evaluate the e predictive effect of the model. The Bootstrap resampling method was used to verify the model and draw the calibration curve of the prediction model. Results The incidence of frailty was 30.07%. The multiple factor analysis showed that poor health status (OR=70.885)/general health status (OR=18.018), smoking (OR=0.483), social participation (OR=0.169), impaired instrumental ability of daily living (OR=3.131) were independent risk factors for frailty (all P<0.05). The area under the ROC curve of the nomograph prediction model was 0.837 (95% CI: 0.808 ～ 0.881, P<0.001), sensitivity of 0.798, specificity of 0.792; the X² value of Hosmer-Lemeshow was 7.111（P=0.213）. The model validation results showed that the C value was 0.836(95% CI: 0.801~0.873), Brier score was 0.138, the predicted value was consistent with the actual value basically. Conclusion The incidence of frailty in elderly patients with CHD is high, which is affected by many influencing factors, health status, smoking, social participation and ability of daily life have certain predictive value for frailty, that can provide reference for early screening and intervention of elderly patients with coronary heart disease.