Risk factors and development of a predictive model for frailty in patients with heart failure

To identify factors influencing frailty in patients with heart failure (HF) and develop a predictive model for clinical use.

A retrospective analysis was conducted on 350 HF patients at Shanghai Baoshan District Hospital of Integrated Traditional Chinese and Western Medicine between January 2020 and December 2023. Of these, 245 patients were allocated to the modeling group (n = 245) and 105 to the validation group (n = 105). In the modeling group, 135 patients were frail and 110 were non-frail. In the validation group, 47 patients were frail and 58 were non-frail. Logistic regression analysis was used to identify factors associated with frailty, and a nomogram was developed and validated to predict frailty risk.

Multivariate logistic regression analysis identified the following independent risk factors for frailty: fall history (OR: 0.101, 95% CI: 0.043-0.242, P < 0.001), advanced age (OR: 0.877, 95% CI: 0.828-0.928, P < 0.001), female sex (OR: 2.925, 95% CI: 1.294-6.613, P = 0.010), low hemoglobin levels (< 12 g/dL; OR: 2.547, 95% CI: 1.816-3.573, P < 0.001), and diabetes (OR: 3.202, 95% CI: 1.559-6.577, P = 0.002). Using these five variables, a nomogram was constructed to predict frailty risk, demonstrating an AUC of 0.822 (95% CI: 0.771-0.907).

Fall history, advanced age, female sex, low hemoglobin levels, and diabetes are significant independent risk factors for frailty in HF patients. The nomogram prediction model demonstrated strong predictive performance, with high accuracy and clinical applicability.